Deformation Mechanisms, Rheology and Tectonics" from Minerals to the Lithosphere
Geological Society Special Publications Society Book Editors R. J. PANKHURST (CHIEF EDITOR) P. DOYLE F. J. GREGORY J. S. GRIFFITHS A. J. HARTLEY R. E. HOLDSWORTH J. A. HOWE P. T. LEAT A. C. MORTON N. S. ROBINS J. P. TURNER
Special Publication reviewing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society's Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society has a team of Book Editors (listed above) who ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Society. The referees' forms and comments must be available to the Society's Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. Geological Society Special Publications are included in the ISI Index of Scientific Book Contents, but they do not have an impact factor, the latter being applicable only to journals. More information about submitting a proposal and producing a Special Publication can be found on the Society's web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways: GAPmS, D., BRUN, J. P. & COBBOLD, P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243. KOEHN, D, ARNOLD, J. & PASSCHIER, C. W. 2005. Fracture and vein patterns as indicators of deformation history, a numerical study. In: GAPAIS, D., BRUN, J. P. & COBBOLD, P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 11-24.
G E O L O G I C A L SOCIETY SPECIAL PUBLICATION NO. 243
Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere EDITED BY
D.
GAPAIS,J. P. BRuN and P. R. COBBOLD Universite de Rennes, France
2005 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society's fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society's international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the American Geological Institute (AGI), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists' Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies' publications at a discount. The Society's online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London WlJ 0BG: Tel. +44 (0)20 7434 9944; Fax +44 (0)20 7439 8975; E-mail: enquiries @geolsoc.org.uk. For information about the Society's meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society's Corporate Affiliates Scheme, write to
[email protected].
Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BAI 3JN, UK (Orders: Tel. +44 (0)1225 445046 Fax +44 (0)1225 442836) Online bookshop: www.geolsoc.org.uk/bookshop The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. 9 The Geological Society of London 2005. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/05/$15.00.
British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. I S B N 1-86239-176-9 Typeset by Techset Composition, Salisbury, UK Printed by Alden Press, Osney Mead, UK
Distributors USA AAPG Bookstore PO Box 979 Tulsa OK 74101-0979 USA Orders: Tel. +1 918 584-2555 Fax +1 918 560-2652 E-mail
[email protected] India Affiliated East-West Press Private Ltd Marketing Division G-l/16 Ansari Road, Darya Ganj New Delhi 110 002 India Orders: Tel. +91 11 2327-9113/2326-4180 Fax +91 11 2326-0538 E-mail affiliat @vsnl.com Japan Kanda Book Trading Company Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-shi, Tokyo 206-0034 Japan Orders: Tel. +81 (0)423 57-7650 Fax +81 (0)423 57-7651 E-mail geokanda@ ma.kcom.ne.jp
Contents
Preface Dedication
REDDY, S. M. & BUCHAN,C. Constraining kinematic rotation axes in high-strain
vii ix 1
zones: a potential microstructural method?
KOEHN, D., ARNOLD,J. & PASSCHIER,C. W. Fracture and vein patterns as
11
indicators of deformation history: a numerical study
MIRABELLA,F., BOCCALI,V. & BARCHI,M. R. Segmentation and interaction of
25
normal faults within the Colfiorito fault system (central Italy)
AUSTIN, N. J. & KENNEDY,L. A. Textural controls on the brittle deformation of
37
dolomite: variations in peak strength AUSTIN, N. J., KENNEDY,L. A., LOGAN,J. M. & RODWAY,R. Textural controls on the brittle deformation of dolomite: the transition from brittle faulting to cataclastic flow
51
RENARD, F., ANDRI~ANI,M., BOULLIER,A.-M. & LABAUME,P. Crack-seal patterns: records of uncorrelated stress release variations in crustal rocks
67
ZUBTSOV, S., RENARD, F., GRATIER, J.-P., DYSTHE, D. K. & TRASKINE, V. Single-contact pressure solution creep on calcite monocrystals
81
BARATOUX, L., SCHULMANN,K., ULRICH, S. 8~ LEXA, O. Contrasting microstructures and deformation mechanisms in metagabbro mylonites contemporaneously deformed under different temperatures (c. 650 ~ and c. 750 ~
97
GUEYDAN,F., MEHL, C. & PARRA,T. Stress-strain rate history of a midcrustal
127
shear zone and the onset of brittle deformation inferred from quartz recrystallized grain size DRURY, M. R. Dynamic recrystallization and strain softening of olivine aggregates in the laboratory and the lithosphere
143
RANALLI, G., MARTIN, S. & MAHATSENTE, R. Continental subduction and exhumation: an example from the Ulten Unit, Tonale Nappe, Eastern Austroalpine
159
RAIMBOURG,H., JOLIVET,L., LABROUSSE,L., LEROY,Y. & AVIGAD,D. Kinematics
175
of syn-eclogite deformation in the Bergen Arcs, Norway: implications for exhumation mechanisms BLENKINSOP, T. G. & KISTERS, A. F. M. Steep extrusion of late Archaean granulites in the Northern Marginal Zone, Zimbabwe: evidence for secular change in orogenic style
193
vi
CONTENTS
BROWN, M. Synergistic effects of melting and deformation: an example from the Variscan belt, western France
205
SPALLA,M. I., ZUCALI, M., DI PAOLA,S. t~ GOSSO, G. A critical assessment of the tectono-thermal memory of rocks and definition of tectono-metamorphic units: evidence from fabric and degree of metamorphic transformations
227
HANDY, M. R. BABIST, J., WAGNER, R., ROSENBERG, C. & KONRAD, M. Decoupling and its relation to strain partitioning in continental lithosphere: insight from the Periadriatic fault system (European Alps)
249
WILLINGSHOFER,E., SOKOUTIS,D. & BURG, J.-P. Lithospheric-scale analogue
277
modelling of collision zones with a pre-existing weak zone
DELACOU,B., SUE, C., CHAMPAGNAC,J.-D. & BURKHARD,M. Origin of the
295
current stress field in the western/central Alps: role of gravitational re-equilibration constrained by numerical modelling Index
311
Preface
This volume is derived from the 13th meeting on Deformation Mechanisms, Rheology and Tectonics (DRT2003). The meeting was held in St Malo (Brittany, France) in April 2003, and organized by an informal group from Gdosciences Rennes (UMR 6118 CNRS, Rennes University), including Michel Ballbvre, St6phane Bonnet, Arlette Falaise, Olivier Galland, Fr6d6ric Gueydan, Charles Gumiaux, Benjamin Le Bayon, Alain-Herv6 Le Gall, Monique Le Moigne, Sylvie Schueller, and C61ine Tirel. It was sponsored by the Centre National de la Recherche Scientifique, Rennes University, the city of Rennes, the Conseil G6n6ral d'Ille et Vilaine, and the R6gion Bretagne. Forty-eight reviewers have worked hard to improve the papers. We thank these persons and institutions for their contributions. The volume contains 18 papers that cover most of the topics and ideas presented and discussed during the meeting. The main approaches are experimental rock deformation, microstructural analysis, field studies, and analogue and numerical modelling. Several papers provide new insights on grainscale and aggregate-scale mechanisms, with new methodological implications (Gueydan et al., Reddy & Buchan, Zubstov et al.) and new advances on the knowledge of deformation processes and rock rheology (Austin & Kennedy, Austin et al., Baratoux et aL, Drury, Gueydan et al., Kiihn et al., Mirabella et al.,
Renard et al., Zubstov et al.) About half of the contributions provide new information or models for processes at crustal or lithospheric scales (Blenkinsop & Kisters, Brown, Delacou et al., Drury, Handy et al., Raimbourg et al., Ranalli et al., Spalla et al., Willingshofer et ai.), with particular emphasis on mantle rheology (Drury), thickening tectonics (Wiilingshofer et aL, Handy et al.) and various aspects of exhumation processes (Blenkinsop & Kisters, Delaeou et al., Raimbourg et al., Ranalli et al. ). The first meeting that led to the DRT series was in Leiden in 1976. Henk Zwart, Richard Lisle, Gordon Lister and Paul Williams were the organizers. The meeting was basically dedicated to Fabrics, Microtextures and Microtectonics. Since then, DRT meetings have accompanied international advances in the understanding of kinematics, mechanics and large-scale tectonic processes, which largely derived from important progress in the knowledge of rock rheology via experimental deformation, and in physical and numerical modelling. We hope that the present volume will further contribute to strengthening the links between studies of geological deformation processes at microscopic to lithospheric scales. Denis Gapais, Jean-Pierre Brun and Peter Cobbold Rennes, October 2004
Dedication to Pierre Choukroune
Pierre Choukroune was born in Casablanca on 28 March 1943. He celebrated his 60th birthday a few weeks before the DRT2003 meeting in St Malo. Pierre received a PhD from Paris University in 1967 and a 'Doctorat d'Etat' from Montpellier University in 1974. The subject of both theses was the structural geology of the Pyrenees. The second thesis was published as Mrmoire 127 of the Socirt6 Grologique de France. Pierre started his professional career in 1967 as 'Assistant' (Assistant Lecturer) at Montpellier University. In 1975, he moved to Rennes as Professor. He stayed at Rennes until 1995, when he moved to the University of AixMarseille. Building on his early fieldwork in the Pyrenees, Pierre rapidly acquired a prominent and highly personal scientific profile. His interests have always spanned various scales, from detailed analysis of microstructures, through regional patterns of strain and displacement (e.g. Choukroune 1976) to plate tectonics (Choukroune 1992). Of the more than 100 papers that he has published in international journals over the last three decades, many have had major impacts on structural geology and tectonics. One of his first papers (Choukroune 1971) was on the analysis of pressure shadows around pyrite crystals, as a tool to understand the development of slaty cleavage. As pointed out by Ramsay
and Hubert (1983) in their textbook, 'This is an outstanding paper from the view point of descriptive excellence, the quality of the diagrams and photographs and the theoretical analysis of data'. This paper contains many of the ideas that Pierre further put forward, in particular the search for criteria of shear senses in deformed rocks. Soon after his arrival in Rennes, Pierre was taken to see the granite mylonites of the South Armorican Shear Zone. He immediately realized the significance of the typical fabric of the rock that results from the coeval development of a schistosity and of shear bands parallel to the bulk shearing plane, and consequently introduced the now famous concept of C/S fabrics. Pierre put every effort into bringing them to international attention (Bertha et al. 1979a, b). The discovery was to open more than a decade of research on shear criteria. Because of his wish to find a strong link between plate tectonics and structural geology, Pierre has been in close contact with a number of geophysicists. In 1974, Xavier Le Pichon asked him to join FAMOUS, the French-American project that studied fault patterns on the Mid Atlantic Ridge (Choukroune et al. 1978). He later took his expertise to the East Pacific Rise (Hekinian et al. 1983; Choukroune et al. 1985) and the Gulf of Aden, where he acted as Chief Scientist for the CYADEN cruise (Choukroune et al. 1988). When the French deep seismic program ECORS was launched in the early 1980s, Pierre acted as Chief Scientist for the first French-Spanish profile across the eastern Pyrenees. The result was a major contribution of deep seismic data to the understanding of a mountain belt (Choukroune et al. 1989). As a team leader, Pierre had a profound influence on the development of structural geology and tectonics at G~osciences Rennes. His initiatives in the Hercynides, the Alps (Choukroune et al. 1986), and several Archaean cratons (Choukroune et al. 1995; Choukroune & Ludden 1997) provided training grounds for a number of young structural geologists from Rennes and elsewhere. None of his PhD students and colleagues at Rennes escaped his thoughtful suggestions and his unusual scientific lucidity and inspiration. Even though he liked to say, 'we
x
DEDICATION TO PIERRE CHOUKROUNE
are a team', what some researchers call the 'Rennes school' bears his signature. As a professor, Pierre's teaching style - elegant and to the point - has inspired many careers in Earth Sciences. His book, 'D6formations et d6placements dans la crofite terrestre' (1995), summarizes much of his scientific philosophy. As well as an outstanding scientist and teacher, Pierre has been an efficient and distinguished academic administrator, for the University, the CNRS, and the Ministry of Education, where he has held a number of important positions. The DRT meeting in St Malo was shortly after his 60th birthday. W e therefore take the greatest pleasure in dedicating the proceedings to Pierre Choukroune, a talented, innovative and inspiring Earth scientist. Jean-Pierre Brun, Peter Cobbold and Denis Gapais
References BERTHt~, D., CHOUKROUNE,P. & Jt~GOUZO,P. 1979a. Orthogneiss, Mylonite and non-coaxial deformation of granites: the example of the South Armorican Shear Zone. Journal of Structural Geology, 1, 31-42. BERTHI~, D., CHOUKROUNE,P. &; GAPAIS, D. 1979b. Orientations pr~f&rentielles du Quartz et orthogneissification progressive en r~gime cisaillant: exemple du Cisaillement Sud-Armoricain. Bulletin de Mindralogie, 102, 265-272. CHOUKROUNE, P. 1971. Contribution ~ l'6tude des m6canismes de d6formation avec schistosit6 grfice aux cristallisations syncin6matiques dans les 'zones abrit6es' ('pressure shadows'). Bulletin de la Socidtd Gdologique de France, 13, 257-271. CHOUKROUNE, P. 1976. Strain patterns in the Pyrenean Chain. Philosophical Transactions of the Royal Society of London, 283, 271-280.
CHOUKROUNE, P. 1992. Tectonic evolution of the Pyr6nEes. Annual Revues of Earth and Planetary Sciences, 20, 143-158. CHOUKROUNE, P. 1995. Ddformations et Ddplacements dans la Cro~te Terrestre. Masson, Paris. CHOUKROUNE,P. & Ecors team. 1989. The Ecors deep seismic profile reflection data and the overall structure of an orogenic belt. Tectonics, 8, 23-39. CHOUKROUNE, P. & LUDDEN, J. N. 1997. Archaean Crustal Growth and Tectonic Processes. Geological Society, London, Special Publications, 121, 63 -98. CHOUKROUNE, P., FRANCHETEAU,J. & Le PICHON, X. 1978. In situ observations along Transform Fault 'A' near 37~ in the FAMOUS area, Mid-Atlantic Ridge. Geological Society of America Bulletin, 89, 1013-1029. CHOUKROUNE, P., FRANCHETEAU,J. & HEKINIAN,R. 1985. Carte g6ologique de la Ride Est Pacifique Bulletin de la Socidt~ Gdologique de 12~ France, 1, 145-148. CHOUKROUNE, P., FRANCHETEAU, J. et al. 1988. Tectonics of an incipient oceanic rift. Marine Geophysical Research, 9, 147-163. CHOUKROUNE, P., BALLI~VRE, M., COBBOLD, P. R., GAUTHIER, Y., MERLE, 0. & VUICHARD, J. P. 1986. Deformation and motion in the Western Alpine Arc. Tectonics, 5, 215-226. CHOUKROUNE, P., BOUHALLIER, H. & ARNDT, N. 1995. Soft Archean lithosphere during periods of crustal growth or reworking. Early Precambrian Processes. Geological Society, London, Special Publications, 95, 67-86. HEKINIAN, R., FRANCHETEAU, J., RENARD, V., BALLARD, R. D. & CHOUKROUNE, P. 1983. Intense hydrothermal activity at the axis of the East Pacific Rise near 13~ Submersible witnesses the growth of sulphide chimney. Marine Geophysical Research, 6, 1-14. RAMSAY, J. G. & HUBERT, M. I. 1983. The Techniques of Modern Structural Geology, Volume 1: Strain Analysis. Academic Press, London.
Constraining kinematic rotation axes in high-strain zones: a potential microstructural method? S T E V E N M. R E D D Y & C R A I G B U C H A N Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia (e-mail: S. Reddy @ cu rtin. edu. a u / C. Buchan @ curtin, edu. au ) Abstract: The correct determination of the kinematic rotation axis in high-strain zones is
essential to the study of the tectonic evolution of the Earth's crust. However, the common assumption that the kinematic rotation axis lies orthogonal to the XZ plane of the finite strain ellipse may be invalid in the case of general shear. Orientation data obtained by electron backscatter diffraction from calcite, deformed in the high-strain Gressoney Shear Zone of the Western Alps, has been investigated using orientation maps, bulk sample crystallographic orientation and misorientation analyses, and detailed intragrain misorientation and crystallographic dispersion analysis. The results demonstrate a strong geometrical coincidence amongst (1) the bulk macroscopic kinematic rotation axis, (2) the orientation of misorientation axes associated with low-angle boundaries, and (3) rotation axes associated with crystallographic dispersion at the intragrain scale. This coincidence is interpreted to reflect a geometric control of the kinematic framework of the high-strain zone on the activity of crystal slip systems. It is proposed that this relationship may be exploited as a new microstructural tool to determine the orientation of bulk kinematic rotation axes in high-strain zones without assuming a geometric link between kinematic rotation and XZ sections. Although further testing is required, application of the approach may lead to a significant advance in our understanding of natural general shear deformation.
Understanding the kinematic evolution of highstrain zones (HSZ) is essential for interpreting the tectonic evolution and palaeogeographic reconstruction of modern and ancient orogens. Commonly HSZ are assumed to have developed by simple shear, enabling mineral lineation orientations to be geometrically related to the transport direction (Ramsay 1980; Hanmer & Passchier 1991). However, theoretical modelling of 'general shear', involving simultaneous components of coaxial and non-coaxial shear, predicts patterns of deformation that are far more complex than those produced by end member pure and simple shear alone (Sanderson & Marchini 1984; Fossen & Tikoff 1993, 1998; Robin & Cruden 1994; Tikoff & Teyssier 1994; Jones et al. 1997; Tikoff & Greene 1997; Jiang & Williams 1998; Lin et al. 1998; Passchier 1998). One important complexity is that the kinematic rotation axis (or vorticity vector) of the deformation may not lie orthogonal to the orientation of maximum finite stretch indicated by mineral stretching lineations. Consequently, the XZ surfaces of HSZ may not be the correct surface from which to gather kinematic (shear sense) data.
Despite the importance of constraining tectonic transport direction, there are currently few criteria with which to quantify the orientation of the kinematic rotation axis that are independent of the assumption that mineral stretching lineations lie orthogonal to the kinematic rotation axis. This paper presents quantitative microstructural data obtained by electron backscatter diffraction (EBSD) of polycrystals defornaed in an HSZ. The data demonstrate a geometric correlation between macroscopic kinematic rotation axes and both misorientation and dispersion axes associated with intragrain low-angle boundaries at the single and multiple grain scale. These observations indicate a potential means of constraining kinematic rotation axes that is independent of the orientation of mineral stretching lineations. The data presented here come from a single calcschist sample ($3-74 located at N45~ ' E7~ ') taken from the Gressoney Shear Zone of the Western Italian Alps (Fig. 1), an HSZ documented in some detail in previous studies (Reddy et al. 1999, 2003; Wheeler et al. 2001a). The Gressoney Shear Zone is dominated by bulk top-to-SE extensional deformation that
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 1-10. 0305-8719/05/$15.00 9 The Geological Societv of London 2005.
2
S.M. REDDY & C. BUCHAN
Fig. 1. Geological maps of the internal zones of the Alps (modified after Reddy et aL 2003). (a) Simplified geological map of the Western Alps; location of (b) marked by rectangular box. (b) Geological map of the internal zones of the Western Alps north of the Aosta Fault showing location of sample $3-74. Arrows indicate sense of shear with relative displacement of the hangingwall marked by the arrowhead. EMC, Eclogitic Micaschist Complex; GMC, Gneiss Minuti Complex; IIDK, Seconda Zona Diorito Kinzigitica;MR, Monte Rosa; AS, Arolla Schist; VP, Valpelline; PK, Pillonet Klippe. The heavy line indicates the contact between oceanic eclogite- and greenschist-facies rocks that mark the base of the Gressoney Shear Zone. was responsible for exhumation of oceanic eclogite facies rocks from lower to mid-crustal levels between 45 and 36 Ma ago (Reddy et al. 1999, 2003). Conservative estimates of bulk strain indicate "g > 50 across the 1 km wide HSZ (Wheeler et al. 2001a). However this strain was spatially and temporally localized (Reddy et aI. 2003) and in detail the Gressoney Shear Zone has a complex deformation history that may include kinematic partitioning (Reddy et al. 1999). The analysed sample contains well-developed kinematic indicators, including shear bands and mica fish structures that could be observed in three dimensions in the field. At both the macro- and microscopic scales, mineral stretching lineations can be traced continuously into shear bands suggesting that the shear bands are a true reflection of the overall sense of shear (Reddy et al. 1999, 2003). The symmetry plane of these monoclinic fabric elements contains the mineral elongation, and the pole of this plane is interpreted to represent the kinematic rotation axis. This geometry indicates either deformation approximating simple shear or general shear deformation in which the coaxial stretching direction was subparallel to the
non-coaxial maximum stretch component (Type E transpression of Fossen & Tikoff 1998). Since Gressoney Shear Zone deformation is not consistent with significant constrictional strain (Reddy et al. 1999, 2003) the former option is more likely. To summarize for sample $3-74, the kinematic rotation axis lies in the foliation plane, close to 90 ~ from the mineral lineation and therefore subparallel to Y (Fig. 2), and the deformation is close to end-member simple shear.
Analytical procedure An oriented petrographic thin section of the sample, cut parallel to the XZ section (XZ = 130.58NE, X = 16.321) was polished with progressively finer grades of diamond paste down to a 0.25 txm diamond polish. The section was then polished using a 0.06 txm colloidal silica suspension in a NaOH solution of pH 10 on a Buehler Vibromet II T M polisher. The sample was studied optically before being analysed in the scanning electron microscope (SEM). For EBSD analysis, the thin section was mounted in a Phillips XL30 SEM (operating conditions were 20 kV and spot size c. 0.7 txm) and tilted
CONSTRAINING KINEMATIC ROTATION AXES IN HIGH-STRAIN ZONES Kinematic Rotation
~ ........
!
' |
g ~
Fig. 2. Idealized HSZ coordinate reference frame used throughout this paper. The XY plane marks the shear plane of the HSZ with X the shear direction. In an HSZ deforming by simple shear the kinematic rotation axis corresponds to Y. Z marks the pole to the HSZ.
to 70 ~ Electron backscatter diffraction patterns (EBSPs) were obtained using an HKL Technology Nordlys detector and data were collected and processed using HKL Technology's Channel 5 software. Each EBSP was automatically matched by the Channel 5 software to theoretical EBSPs for calcite to give orientation information for each grid node. The closeness of fit between observed and theoretical patterns was good (average mean angular deviation = 0.62). Data were noise reduced using the Channel 5 'wildspike' correction and a four-neighbour zero solution extrapolation. For this study a single orientation map was constructed by automatically collecting EBSPs from a user-defined grid of 120 x 70 points at 4 txm spacing (map area = 480 x 280 txm). The average grain diameter of the mapped area was 17 ~m (range = 5-108 ~m) with mean grain area of 347 txm2. Zero solutions making up 18% of the map mostly reflect minor quartz and white mica in the sample. Channel 5 software was used to produce orientation maps, and calculate the minimum misorientation angle and misorientation axis geometry. The crystal symmetry of calcite means that a number of possible misorientation angles and axes may be calculated between two differently oriented crystals or parts of crystals. However, following several other workers, the angle/axis pair corresponding to the minimum misorientation angle was used for misorientation analysis (Pospiech et al. 1986; Mainprice et al. 1993; Lloyd et al. 1997; Wheeler et al. 2001b). This is justified for intragrain low-angle misorientation analysis.
3
Results The results from a single orientation map (Fig. 3a) indicate a strong crystallographic preferred orientation (CPO) for most of the calcite grains analysed (cf. Fig. 3a, d). The orientation of the c-axis cluster is consistent with macroscopic kinematic indicators that record topto-SE shearing (Fig. 3d). The only volumetrically significant exception to the CPO is grain 1 which has an a-pole parallel to the dominant c-direction in the sample (Fig. 3d). The uncorrelated misorientation angle distribution (Fig. 4a) indicates a smaller number of high misorientation angles than that predicted for a random distribution and is consistent with the presence of a CPO. The absence of a misorientation angle peak in the correlated data at 78 ~ indicates that few of the calcite grains are twinned (e.g. Bestmann & Prior 2003). However, twinning is present in other parts of the sample. The correlated misorientation angle distribution form the whole mapped area indicates numerous (c. 2000) low-angle boundaries between 2 ~ and 10 ~ (Figs 3a and 4a). These have orientations that cluster around the mesoscopic kinematic rotation axis (Y direction) of the sample (Fig. 4b). Although there are significant errors (up to 30 ~) associated with the determination of misorientations axes for small numbers of low-angle misorientations, these can be minimized by using a large dataset (Prior 1999). The clustering of c. 2000 low-angle misorientation axes around the centre of the pole figure (Fig. 4b) is statistically significant and is likely to reflect a geological process. Misorientation axes associated with correlated misorientation angles > 1 0 ~ have a random distribution (not shown), indicating no systematic relationship between the orientations of different calcite grains. The qualitative distribution of low-angle boundaries (Fig. 3a) is shown in more detail for two grains (1 & 2) in which the orientation data has been coloured to demonstrate crystallographic orientation variations of 15 ~ and 12 ~ respectively from the (blue) centre of each grain (Fig. 3b, c). These data indicate that the variations in orientation across these grains are systematic, forming discrete bands of constant relative misorientation towards the grain edges (Fig. 3b, c). The overall bending of the crystal lattice is cumulative and there are no individual angular misorientations greater than 10 ~ within either of the two grains. The systematic reorientation of crystal axes by low-angle boundaries is also apparent in pole figures of crystallographic orientation for grains 1 and 2 (Fig. 5) in which colour variations in the pole figures correspond
4
S.M. REDDY & C. BUCHAN
Fig. 3. (a) Automated EBSD map of calcite in $3-74 XZ (130.58NE) section. Each pixel (total = 6847) represents an individual orientation analysis. Colonrs reflect variations in orientation defined by the three Euler angles. Red and black lines represent boundaries with >2 and > 10~ misorientation respectively between adjacent analyses. Black pixels show analyses of quartz or white mica (data not shown) or zero solutions. Grains 1 and 2 correspond to grains shown in (b) and (c) respectively. (b) & (e) Detail of grains 1 (475 points) and 2 (304 points) derived by assigning a spectrum of colour (blue to red) to a range of 15~ and 12~ misorientations respectively from the centre of each grain. Systematic changes in colour indicate orientation variations due to low-angle boundaries. The spatial distribution of colours can be directly compared to crystallographic orientations indicated in Fig. 5. (d) Lower hemisphere, equal angle projections of calcite orientation data (c-, a- and r-poles) illustrated in (a). X and Z correspond to shear zone coordinates shown in Fig. 2. Colours in (a) correspond directly to position of orientation data in (d). Orientation data from grains 1 and 2 are indicated.
to the colours in the respective orientation maps (Fig. 3b, c). The crystallographic axes within two grains of different bulk crystallographic orientation (grains 1 & 2) are dominantly dispersed around small circles centred on the centre of the pole figures, that is, the macroscopic Y direction of the shear zone coordinate system (Fig. 5a, b). In both grains, one crystallographic axis at the centre of the pole figures shows little dispersion (grain 1 = r-pole; grain 2 -- a-pole). The relative lack of dispersion of this crystallographic orientation indicates that this represents a principal
rotation axis associated with low-angle boundary formation within the grains. In detail there are subtle complexities to this simple picture. In grain 1 there is slight dispersion of the central r-pole and other crystallographic axes that indicate minor dispersions away from a single small circle distribution centred around Y. The exact orientation of this second rotation axis is difficult to constrain due to the limited total angular dispersions associated with this axis ( < 5~ H o w e v e r it is interpreted to lie in the lower left quadrant of the pole figure. In grain 2, a second minor dispersion seemingly
CONSTRAINING KINEMATIC ROTATION AXES IN HIGH-STRAIN ZONES
Misorientation Distribution
5
coordinate system (Fig. 6c, d, g, h) indicate very little preferred orientation.
Discussion
,
Calcite deformation and the role of different slip systems
o 0o
10
20
h'~
Z
30
40 50 60 70 Misodentation Angle (~
80
90
100
Z
-
Fig. 4. (a) Minimum misorientation angle distribution for orientation data indicated in Fig. 3a. White histogram corresponds to neighbour-pair misorientations (correlated data), the black histogram representing random-pair misorientations (uncorrelated data). Uncorrelated data show a subsample of 2000 randomly chosen random-pair misorientations. Solid black line indicates the theoretical misorientation angle distribution of randomly oriented trigonal minerals. (b) The distribution of misorientation axes for minimum correlated misorientation angles of 2-5 ~ and 5-10 ~ shown in (a). These data indicate a concentration of low-angle misorientations axes around the sample Y direction. around small circles centred around the f-pole in the lower right quadrant accommodates a few degrees of rotation. In both cases these additional rotational components are minor and only slightly modify the dominant dispersion centred around the sample Y direction, that is, the r-pole and a-pole for grain 1 and 2 respectively. Misorientation axes for 2 - 5 ~ and 5 - 1 0 ~ minimum correlated misorientation angles plotted with respect to sample coordinates for grains 1 & 2 show a relatively widespread distribution (Fig. 6a, b, e, f). This is particularly evident for the 2 - 5 ~ misorientation data. Such a result is expected because of large errors involved with calculating misorientation axes associated with low-angle misorientations (Prior 1999). However, in general the average distribution of misorientation axes is again close to the Y direction of the shear zone coordinate system (Fig. 2). The inverse pole figures showing the position of misorientations axes with respect to the crystal
In two grains of differing orientation, the dispersion of crystallographic axes associated with low-angle boundaries appears to be dominated by rotation about an axis that lies parallel to the kinematic Y direction of the sample (Fig. 5). For grain 1, the rotation axis associated with the dispersion of crystallographic axes has an orientation that corresponds to an r-pole (Fig. 5). The dominance of a single dispersion axis for this grain suggests that most deformation was accommodated on low-angle boundaries developed by activity of a single crystallographic slip system. However, the obliquity of lowangle boundary traces on the orientation map (Fig. 3b), and the presence of a second minor dispersion axis (Fig. 5a), indicates the presence of at least one other slip system. Similarly, low-angle boundaries in grain 2 (Fig. 5) also developed by multiple slip systems but were dominated by dispersion around an axis parallel to one of the a-poles. The correlation of axes associated with dominant crystallographic pole dispersion with different crystallographic directions in grains 1 and 2 (r-pole and a-pole respectively) indicates that the dominant slip systems were different in the two grains. Crystallographic pole dispersion patterns can be used to identify intragrain slip systems using simple models of tilt boundary formation during the formation of edge dislocations (Lloyd & Freeman 1991, 1994). For tilt boundaries, the rotation axis for crystal slip is the normal to the plane that contains the pole to the slip plane and the slip direction, and therefore the three features are mutually orthogonal in calcite (De Bresser 1991; Bestmann & Prior 2003). Given that the dispersion axis lies in the centre of the pole figure, the slip plane and slip direction responsible for dispersion must lie on the primitive circle of the pole figure. Using this model, the data from grain 1 indicate the probability of a-slip ((-12-10) (-2021)) as the active slip system. Experimental evidence for a-slip in calcite is rare (Turner & Heard 1965; Paterson & Turner 1970) and consequently is largely ignored as an important slip system (e.g. De Bresser & Spiers 1997). However, no other published calcite slip systems readily account for the observed dispersion around the r-pole (Fig. 5a). In addition, Schmid Factor
6
S . M . REDDY & C. B U C H A N
a) Grain 1
Z
{-~olz}
b) Grain 2
y
Z
X
c Fig. 5. Equal area, lower hemisphere projections of orientation data (poles to c, r, f, e, a and direction (-2021)) for grains 1 and 2 of Fig. 3. In both grains most orientation data are dispersed in small circles around the centre of the pole figures. Dispersion axes are coincident with the r- and a-pole in grain 1 and 2, respectively. Minor secondary dispersion in both grains can also be identified and relates to the operation of secondary minor slip systems that cumulatively accommodate a few degrees of rotation,
CONSTRAINING KINEMATIC ROTATION AXES IN HIGH-STRAIN ZONES z
Grain 1
Grain 1
\
. •
7
",4
..
""" "
f
9
9
"" i;,.'. :"'_-" ,.-,. ".~)" ",
c) m z
9.
~t
Grain2
":'" "-
~''~'.,
=
x .......................
~
~
~
" -.
Grain 2
"': "" ~'" ...... = u ;#
g)
,
.
"=
=,
,~ |
II
....-',....
." ~""
h)
Fig. 6. Equal area, lower hemisphere projections of minimum misorientation axes orientation for low-angle boundaries in grains 1 and 2 plotted in both sample coordinate (pole figure) and crystal coordinate (inverse pole figure) space for misorientation angles of 2 - 5 ~ and 5-10 ~
analysis, assuming deformation was by simple shear, indicates higher resolved shear stress associated with a-slip (Schmid Factor = 0.5) than any other calcite slip system, further supporting the possibility of a-slip in grain 1. Since grain 1 is anomalous in its orientation (Fig. 3a, d) and makes up less than 5% of the analysed grains, its contribution to deformation of the sample as a whole is minor. The additional secondary dispersion axis identified in the pole figures (Fig. 5a) cannot be readily identified because of its small angular misorientations. Grain 2 has an orientation similar to the majority of analysed grains and records dispersion about the central a-pole. This dispersion axis is common to both of the common calcite slip systems r- and f-slip. However, only in the case of r-slip is the slip direction also lying on the primitive circle at 90 ~ to the pole to the slip plane (cf. Bestmann & Prior 2003). This geometry discriminates between the two slip systems and indicates the probability that the major dispersion in grain 2 was accommodated by r-slip ((10-14) (-2021)). Importantly, despite the different slip systems, the rotation axis associated with low-angle boundary formation in both grains lies parallel to the macroscopic kinematic rotation axis (Y), suggesting a kinematic control on the activity of the dominant slip systems.
The recognition of specific crystallographic dispersion axes for grains 1 and 2 should also be recorded by the geometry of low-angle misorientation axes both in sample and crystal coordinate systems (Fig. 6). However, the misorientation data are not easily reconciled with the crystallographic pole dispersion data. Despite having an average orientation close to the Y axis, misorientation axes plotted in the sample coordinate system have a distribution that demonstrates a large degree of variation, particularly in the lower (2-5 ~ misorientations angles (Fig. 6a, b, e, f). The apparent almost random distribution of misorientation axes in the crystal coordinate system (Fig. 6c, d, g, h) is also inconsistent with the specific crystallographic dispersion axes identified earlier (Fig. 5), which would be expected to cluster around r- and a-poles for grains 1 and 2 respectively. There are several possible reasons for the observed discrepancy between the crystal pole dispersion axes and the misorientation axes. Misorientation axes calculated from small angular misorientations have a significant error (tens of degrees) associated with them (Prior 1999). Such an error is significant in sample coordinate space (Prior 1999) and is amplified in crystal coordinate space where axes are plotted in a smaller angular volume. The error
8
S.M. REDDY & C. BUCHAN
associated with the calculation of misorientation axes decreases with increasing misorientation angle (Prior 1999) and it is noticeable that in grain 1, where we have sufficient 2 - 5 ~ and 5 - 1 0 ~ misorientations to make a comparison, that the distribution of, the misorientation axes for the larger misorientation angles is more tightly clustered around the kinematic Y direction (Fig. 6b). Furthermore, the 5 - 1 0 ~ misorientation axes in crystal coordinate inverse pole figures (Fig. 6d) are less variable than the 2 - 5 ~ axes (Fig. 6c). At least part of the variation in the distribution of misorientation axes is therefore attributed to errors associated with the calculation of misorientation axes for small angular variations. The misorientation axis is the common axis between two different crystallographic orientations around which a rotation through the misorientation angle will bring the two orientations into coincidence (Pospiech et al. 1986). As such it is a geometric construct that need not relate to a specific process. Changes in orientation associated with a number of different slip systems may therefore yield misorientation axes that are a combination of the different slip systems and may not have simple relationships to crystallographic orientations (Lloyd & Freeman 1991). Although in both grains 1 and 2 the dominant slip system is associated with a crystallographic pole dispersion axis centred on the sample Y direction, the presence of other operating slip systems is also apparent from secondary dispersions. In grains 1 and 2, the secondary slip systems are minor and accommodate small misorientations so a combination of different slip systems should be more apparent in the 2 - 5 ~ misorientations than in the larger misorientation angle data. Consequently, some of the variation in the orientation of 2 - 5 ~ misorientation axes may be a result of a combination of minor amounts of slip on different slip systems. As a result of the potential complications and the errors outlined above, individual misorientation axes are considered unreliable for identifying crystal slip systems associated with the development of low-angle boundaries. Dispersion axes obtained from crystallographic pole dispersion patterns (Fig. 5), particularly those illustrating a large cumulative angle of rotation, are considered to be a more robust method. However, the data presented here indicate that a large number of low-angle misorientation axes yield an average orientation that is identical to the rotation axis associated with crystallographic dispersion. Many studies investigating slip systems within deforming minerals have recognized the
importance of critically resolved shear stress (CRSS) for the initiation of slip on a particular slip system (e.g. Lister 1978; Takeshita et al. 1987 for calcite). The CRSS model assumes that slip on a particular system will occur when the shear stress on that slip system reaches a threshold value. Numerical investigations of slip systems in stress space indicate that the relative activity of these systems is controlled by the geometrical relationship of the yield surface to the imposed strain vector rather than the value of the CRSS (Lister & Hobbs 1980; Takeshita et al. 1987) suggesting that there is a geometrical control, in addition to a CRSS control, on slip system activity. The dispersion pattern dominated by a single rotation axis shown by our data suggests that activity is dominated by a single slip system in each grain, with a minor component of at least one other slip system. This is inconsistent with deformation as assumed in the Taylor models that consider single crystal yield surfaces requiring up to five active slip systems (Von Mises criteria) (Takeshita et al. 1987). Taylor modelling ignores the effect of dynamic recrystallization and recovery that takes place in natural deformations, as well as other deformation processes such as diffusive mass transfer, diffusion accommodated grain boundary sliding and grain boundary migration that may relax the necessity for five active slip systems. Experimental studies involving dynamic recrystallization at high temperatures indicate that geometric softening associated with the preferred orientation of specific calcite slip systems may take place and lead to the accommodation of large shear strains (Pieri et al. 2001). Studies of dynamic recrystallization during natural deformation also indicate that within individual grains, deformation may be accommodated by the operation of a single slip system (Bestmann & Prior 2003). The data presented here are consistent with these findings and indicate a similar geometric control on the activity of particular slip systems, whereby the dominant slip system that operates within a particular deforming grain is one in which the geometric rotation axis associated with the slip system lies parallel to the mesoscopic kinematic rotation axis. Microstructural characterization o f the kinematic rotation axis
The data presented in this study demonstrate a geometrical coincidence ofintragrain rotation axes responsible for the dispersion of crystallographic axes (Fig. 5), mean low-angle misorientation axes orientations from single crystals (Fig. 6),
CONSTRAINING KINEMATIC ROTATION AXES IN HIGH-STRAIN ZONES low-angle misorientation axes collected from polycrystals (Fig. 4), and the macroscopic kinematic rotation axis of the HSZ. These observations indicate that rotation and misorientation axes associated with low-angle boundaries may provide geometric constraints on the orientation of the macroscopic kinematic rotation axis. This similarity of misorientation and rotation axes for low-angle boundaries with mesoscopic kinematic rotation axes has been noted previously in EBSD studies (e.g. Bestmann & Prior 2003), although the kinematic significance of this relationship has not been explored. In this study, microscopic analysis has been linked to well-constrained kinematic data in which the shear direction of the studied HSZ has been previously established (Reddy et al. 1999, 2003). In this high-strain example the mineral lineation direction lies orthogonal to the kinematic rotation axis (Fig. 2). However, it cannot be assumed a priori that mineral stretching lineations lie orthogonally to the kinematic rotation axis. For example, in the case of general shear, the maximum finite stretch may be dominated by the coaxial component of the deformation that is independent of kinematic rotation. Despite this, mineral lineations are widely used to constrain shear directions and identify the ideal plane on which kinematic indicators should be viewed (XZ of Fig. 2). This may lead to incorrect interpretations of shear sense in HSZ deforming by general shear. The data presented here indicate that for simple shear deformation the orientation of the kinematic rotation axis may be constrained by identifying the rotation axes associated with crystallographic dispersion across low-angle boundaries and geometrically coincident low-angle misorientation axes. Since crystallographic orientations are independent of the orientation of the analysed section, crystallographic dispersion and misorientation axes are readily identifiable from EBSD data independent of lineation orientation. Although the data presented here refer to conditions approaching simple shear, similar geometrical relationships may exist under general shear conditions. If so, the analysis of misorientation and dispersion axes may yield important information regarding the orientation of kinematic rotation axes in HSZ undergoing general shear. Further detailed studies are required to test this hypothesis.
Microstructural preservation and the temporal evolution of HSZ Owing to the evolving nature of microstructures and their syn- and post-kinematic stability,
9
low-angle boundaries may reflect only the last stages of the deformation history in a particular part of an HSZ. Consequently, kinematic information obtained from the analysis of low-angle boundaries may not represent much of the earlier kinematic evolution of the HSZ and may provide only a snapshot of the lattermost stages of deformation. This presents a problem in constraining variations of kinematic history during the temporal evolution of the HSZ. Recent detailed geochronological studies of the Gressoney Shear Zone indicate that the spatial distribution of deformation with time was heterogeneous, with strain migration and localization leading to the preservation of older parts of the shear zone between zones of younger deformation (Reddy et al. 1999, 2003). Preliminary microstructural data from other Gressoney Shear Zone samples, which record different deformation ages and variable kinematic senses, indicate that the Gressoney Shear Zone was not affected by a homogenous or pervasive, late-stage microstructural overprint and that microstructures (including low-angle boundaries) developed early in the deformation history may be preserved in different temporal domains. Therefore, different parts of the HSZ appear to have a microstructural memory that can be used to constrain the earlier parts of the deformation history. Consequently, the application of the technique presented in this paper to HSZ that record a complex evolution may lead to a significant advance in our understanding of kinematic partitioning in naturally deformed strain systems with respect to both space and time. This work was funded through ARC Large Grant A00106036. The data presented in this paper were collected at the Australian Research Council (ARC) funded Microstructural Analysis Facility (part of the Western Australian Centre of Microscopy)at Curtin Universityof Technology,Perth, Australia.GeoffLloyd,Neil Mancktelow and Denis Gapais are thankedfor comprehensiveand constructivereviews. This paper is TSRC publicationNo. 288 and contributes to the TSRC's Program 4: Tectonic Processes.
References BESTMANN, M. & PRIOR, D. J. 2003. Intragranular dynamic recrystallisation in naturally deformed calcite marble: diffusion accommodated grain boundary sliding as a result of subgrain rotation recrystallization. Journal of Structural Geology, 25, 1597-1613. DE BRESSER,J. H. P. 1991. Intracrystallinedeformation of calcite. Geologica Ultrajectina, 79, 1- 191.
10
S.M. REDDY & C. BUCHAN
DE BRESSER, J. H. P. & SPIERS, C. J. 1997. Strength characteristics of the r, f, and c slip systems in calcite. Tectonophysics, 272, 1-23. FOSSEN, H. & TIKOFF, B. 1993. The deformation matrix for simultaneous simple shearing, pure shearing and volume change, and its application to transpression - transtension tectonics. Journal of Structural Geology, 15, 413-422. FOSSEN, H. & TIKOFF, B. 1998. Extended models of transpression and transtension, and application to tectonic settings. In: HOLDSWORTH, R. E., STRACHAN, R. A. & DEWEY, J. F. (eds) Continental
Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications, 135, 15-33. HANMER, S. & PASSCHIER, C. W. 1991. Shear-sense indicators: a review. Geological Survey of Canada, Paper 90-17, 72. JIANG, D. & WILLIAMS, P. F. 1998. High-strain zones: a unified model. Journal of Structural Geology, 20, 1105-1120. JONES, R. R., HOLDSWORTH, R. E. & BAILEY, W. 1997. Lateral extrusion in transpression zones: the importance of boundary conditions. Journal of Structural Geology, 19, 1201-1217. LIN, S., JIANG, D. & WILLIAMS, P. F. 1998. Transpression (or transtension) zones of triclinic symmetry: natural example and theoretical modelling. In: HOLDSWORTH, R. E., STRACHAN, R. A. & DEWEY, J. F. (eds) Continental Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications, 135, 41-57. LISTER, G. S. 1978. Texture transitions in plastically deformed calcite rocks. Proceedings of the 5th
International Conference on Textures of Materials, 199-210. LISTER, G. S. & HOBBS, B. E. 1980. The simulation of fabric development during plastic deformation and its application to quartzite: the influence of deformation history. Journal of Structural Geology, 2, 355 -370. LLOYD, G. E. & FREEMAN, B. 1991. SEM electron channelling analysis of dynamic recrystallization in a quartz grain. Journal of Structural Geology, 13, 945-953. LLOYD, G. E. & FREEMAN, B. 1994. Dynamic recrystallization of quartz under greenschist facies conditions. Journal of Structural Geology, 16, 867- 881. LLOYD, G. E., FARMER, A. B. & MAINPRICE, D. 1997. Misorientation analysis and the formation and orientation of subgrain and grain boundaries. Tectonophysics, 279, 55-78. MAINPRICE, D., LLOYD, G. E. & CASEY, M. 1993. Individual orientation measurements in quartz polycrystals - advantages and limitations for texture and petrophysical property determinations. Journal of Structural Geology, 15, 1169-1187. PASSCHIER, C. W. 1998. Monoclinic model shear zones. Journal of Structural Geology, 20, 1121-1137.
PATERSON, M. S. & TURNER, F. J. 1970. Experimental deformation of strained calcite crystals in extension. In: PAULITSCH, P. (ed) Experimental and Natural Rock Deformation. Springer, Berlin, 109-141. PIERI, M., KUNZE, K., BURLINI, L., STRETTON, I., OLGAARD, D. L., BURG, J.-P. & WENK, H.-R. 2001. Texture development of calcite by deformation and dynamic recrystallization at 1000 K during torsion experiments of marble to large strains. Tectonophysics, 330, 119-140. POSPIECH, J., SZTWIERTNIA,K. & HAESSNER,F. 1986. The misorientation distribution function. Textures and Microstructures, 6, 201. PR1OR, D. J. 1999. Problems in determining the misorientation axes, for small angular misorientations, using electron backscatter diffraction in the SEM. Journal of Microscopy, 195, 217-225. RAMSAY, J. G. 1980. Shear zone geometry: a review. Journal of Structural Geology, 2, 83-99. REDDY, S. M., WHEELER, J. & CLIFF, R. A. 1999. The geometry and timing of orogenic extension: an example from the Western Italian Alps. Journal of Metamorphic Geology, 17, 573-589. REDDY, S. M., WHEELER, J. et al. 2003. Kinematic reworking and exhumation within the convergent Alpine Orogen. Tectonophysics, 365, 77-102. ROBIN, P. Y. F. & CRUDEN, A. R. 1994. Strain and vorticity patterns in ideally ductile transpression zones. Journal of Structural Geology, 16, 447 -466. SANDERSON, D. J. & MARCHINI, W. R. D. 1984. Transpression. Journal of Structural Geology, 6, 449-458. TAKESHITA, T., TOMI~, C., WENK, H.-R. & KOCKS, U. F. 1987. Single-crystal yield surface for trigonal lattices: Application to texture transitions in calcite polycrystals. Journal of Geophysical Research, 92, 12917-12930. TIKOFF, B. & TEYSSIER, C. 1994. Strain modeling of displacement-field partitioning in transpressional orogens. Journal of Structural Geology, 16, 1575-1588. TIKOFF, B. & GREENE, D. 1997. Stretching lineations in transpressional shear zones: an example from the Sierra Nevada Batholith, California. Journal of Structural Geology, 19, 29-39. TURNER, F. J. & HEARD, H. C. 1965. Deformation in calcite crystals at different srain rates. University
of California Publications in Geological Sciences, 46, 103-126. WHEELER, J., REDDY, S. M. & CLIFF, R. A. 2001a. Kinematic linkage between internal zone extension and shortening in the more external units in the NW Alps. Journal of the Geological Society, London, 158, 439-443. WHEELER, J., PRIOR, D. J., JIANG, Z., SPE1SS, R. & TRIMBY, P. W. 200lb. The petrological significance of misorientations between grains. Contributions to Mineralogy and Petrology, 141, 109-124.
Fracture and vein patterns as indicators of deformation history: a numerical study D A N I E L K O E H N , J O C H E N A R N O L D & CEES W. P A S S C H I E R
Tectonophysics, Institute for Geosciences, Becherweg 21, University of Mainz, 55099 Mainz, Germany (e-mail: koehn @mail. uni-mainz, de) Abstract: Fracture and vein patterns in the brittle crust of the Earth contain information on
the stress and strain field during deformation. Natural examples of fracture and vein patterns can have complex geometries including combinations of extension and conjugate shear fractures. Examples are conjugate joint systems that are oriented with a small angle to the principal stress axis and veins that show an oblique opening direction. We developed a discrete numerical model within the modelling environment 'Elle' to study the progressive development of fractures in two dimensions. Results show that pure shear deformation alone can produce complex patterns with combinations of extension and shear fractures. These patterns change in geometry and spacing depending on the Young's modulus of the deforming aggregate and the initial noise in the system. A complex deformation history, including primary uniaxial loading of the aggregate that is followed by tectonic strain, leads to conjugate shear fractures. During progressive deformation these conjugate shear fractures may accommodate extensional strain or may be followed by a secondary set of extension fractures. The numerical patterns are consistent with joint, fault and vein geometries found in natural examples. The study suggests that fracture patterns can record complex deformation histories that include primary uniaxial loading due to an overlying rock sequence followed by tectonic strain.
Brittle deformation involving fracturing of rocks is a very important deformation mechanism in the upper crust of the Earth (Patterson 1978; Ranalli 1995). Many sedimentary rocks that are exposed today are filled with joints, faults and veins with characteristic patterns reflecting the importance of brittle deformation (Price & Cosgrove 1990). In addition, fracture and vein patterns are used by structural geologists as a map of the far-field stress orientation at the time of fracture formation (Ramsay & Huber 1983; Price & Cosgrove 1990; Oliver & Bons 2001). Fractures are generally classified into mode I, mode II and mode III fractures (Pollard & Segall 1987; Scholz 2002). Mode I fractures are extensional and open perpendicular to the maximum tensile stress. They accommodate strain by opening and may contain vein material that precipitates in the open space. Mode II fractures are shear fractures that develop at an angle to the maximum principal stress. They accommodate strain by slip along the fracture plane and are therefore called shear fractures or faults if significant slip has taken place along them (Bonnet et al. 2001). Mode III fractures are important in
three dimensions and will not be considered further in this paper since the presented models are two-dimensional. In order to understand what kind of fracture will form under a given stress condition and what orientation it will have with respect to one of the major stress axes it is useful to construct a Mohr circle diagram (Jaeger & Cook 1976) with the normal stress on the x-axis and the shear stress on the y-axis (Fig. 1). The Mohr envelope or failure curve in Figure 1 is constructed after Griffith (Griffith 1920; Jaeger & Cook 1976; Scholz 2002). If the Mohr circle crosses the failure curve in the regime of tensile stress, mode I fractures will develop, if it cuts the failure curve in the compressive stress regime, mode II fractures will form with an angle 0 to the direction of the compressive stress. We will term fractures that form by a coalescence of small mode I cracks, extension fractures (joints in a larger volume of rock or veins if they are filled with new material) and fractures that form by a coalescence of mainly mode II cracks, shear fractures (faults in a larger volume of rock if slip takes place). In addition, Hancock (1985) and Price & Cosgrove (1990) mention an intermediate
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. DeformationMechanisms, Rheology and
Tectonics:from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 11-24. 0305-8719/05/$15.00
9 The Geological Society of London 2005.
12
D. KOEHN ET AL.
hybrid i ~ modeII (3"1 , Oli ~ ~
i
failurecurve Ze=4To(O'n+To)
,T , "
,,
, , ..,
", T
0-I (In
', '
~ AO
failure curve
Fig. 1. A Mohr circle diagram with the normal stress (o-.) on the x-axis and the shear stress (~-) on the y-axis. The failure curve is constructed using the Griffith failure criterion (Jaeger & Cook 1976). Two Mohr circles with different sizes are shown. The smaller circle cuts the failure envelope in the tensile regime and produces mode I fractures and hybrid extension/shear fractures. The larger circle cuts the failure envelope in the compressive regime and produces mode II shear fractures. The mean stress (o-m), differential stress (Act) and the two main principal stress components (o'l, o'2) are indicated next to the larger Mohr circle. To represents the tensile strength.
fracture type called hybrid extension/shear fractures that form at an angle to the compressive stress direction but still lie in the tensile stress regime (Fig. 1) and form probably by a coalescence of mode I and II cracks. One has to note, however, that the occurrence of hybrid extension/shear fractures as propagating cracks and their orientation with respect to the principal stress direction and thus their connection to the parabolic failure curve in Figure 1 is questionable (Engelder 1999; Ramsay & Chester 2004). Rocks in the brittle regime can only sustain small amounts of elastic strain before they start to form fractures (Means 1976). If fractures develop, the overall behaviour of the rock will be plastic (Paterson 1978). Fractures accommodate strain in a rock and relieve the overall stress. However, since fractures, are two-dimensional planar structures, in order to accommodate threedimensional strain, complex fracture patterns develop. Anderson (1951) introduced the idea of conjugate faults, which intersect in the intermediate stress direction. The plane perpendicular to the intermediate stress axis then includes the minimum and maximum stress directions and the slip vectors of the faults. This concept, however, only works if the intermediate stress is zero or
close to zero. Otherwise more complex fault or fracture patterns develop with at least four sets of coeval faults, arranged in an orthorhombic symmetry (Oertel 1965; Aydin & Reches 1982; Krantz 1988). Fractures that are filled with vein material pose a problem. Rocks are normally loaded by the overlying rock sequence before they are deformed tectonically, so both the vertical and horizontal stresses will be compressional. Therefore extension fractures are not expected to form since the Mohr circle will cut the failure envelope in the compressive regime (Jaeger & Cook 1976; Means 1976; Suppe 1985; Price & Cosgrove 1990). However, veins are formed quite often in the upper crust, so an alternative mechanism is needed to produce tensile stresses. Two possible sources are (1) a heterogeneous layered sequence where tensile stresses are concentrated in competent layers forming fractureboudinage or (2) the presence of a high fluid pressure that shifts the Mohr circle towards the tensile regime (Price & Cosgrove 1990; Ranalli 1995). In these cases one can find extension fractures that are filled with vein material if material can precipitate. An example of a hybrid shear fracture that is filled with vein material and laterally extends into extension fractures is shown in Figure 2a. An interesting question is if a vein set like the one shown in Figure 2b can also be associated with a single set of conjugate hybrid shear fractures. Examples of conjugate hybrid joint sets can be found in Hancock (1985). In this paper we model the dynamic development of fracture patterns in a two-dimensional approach. We simulate different deformation histories starting from a simple case where we investigate different internal model parameters. Then we extend to more complex deformation histories that include gravitational loading and tectonic strain. We will argue that the initial fracture geometries that later develop into veins contain a record of the gravitational loading that preceded tectonic deformation. It will be shown that even during a simple deformation history different stress states in the samples will lead to different kinds of failures.
The model In order to model two-dimensional fracture patterns we developed a discrete-element model based on the work of Malthe-Screnssen et al. (1998b) and combined it with the modelling environment 'Elle' (Jessell et al. 2001). With this kind of model, linear elastic behaviour of rock aggregates can be simulated with the full description of the strain and stress field. The
FRACTURE AND VEIN PATTERNS
(a)
13
(b)
Fig. 2. (a) Sketch of a vein that may have developed partly as a hybrid shear fracture (after Price & Cosgrove 1990, Fig. 1.63). (b) Bedding plane with conjugate sets of veins from a locality near Sestri Levante in the Liguride units, Italy (see hammer for scale).
model also allows the introduction of fractures by bond-breaking and can thus be used to investigate the behaviour of the aggregate beyond linear elasticity theory. Studies with this kind of model have shown that it reproduces the scaling behaviour and statistics of experimental fracture patterns (Walmann et al. 1996; MaltheS0renssen et al. 1998a, b, 1999) and more complex systems (Jamtveit et al. 2000; Flekk0y et al. 2002). In the discrete model circular particles are connected with their neighbours by linear elastic springs (Fig. 3a). We use a triangular lattice in two dimensions since it reproduces linear elasticity theory with only nearest neighbour interaction (Flekkcy et al. 2002). The force acting on a particle from a neighbour is proportional to the extension or compression of the connecting spring with respect to the equilibrium distance. By definition, compressive stresses in the model are negative and tensile stresses positive. The lattice is in equilibrium condition if all forces acting on a particle cancel out. If the equilibrium condition is perturbed, for example by applying a deformation to the lattice, particles
.(~)~~ ~
(b bound,ary (c)
fracture
Fig. 3. Sketches of the discrete particle model that is used for the numerical simulations. (a) Triangular lattice showing particles and springs that connect particles. (b) Boundary particles (in grey) are used to apply deformation on the modelled aggregate. (c) Fractures are induced by the possibility of breaking springs. Springs break once they have reached a critical tensile stress and are irreversibly removed from the model.
are moved according to resulting forces until all particles are in an equilibrium position defined by a given threshold. In order to find equilibrium for the whole lattice efficiently it is over-relaxed by moving the particles by an 'over-relaxation factor' beyond their equilibrium condition (Allen 1954). In the model, particles fill an initial area called the deformation box. Particles along the box boundaries are defined as wall-particles and are used to apply kinematic boundary conditions (Fig. 3b). These particles are fixed perpendicular to the boundary of the model but can move freely parallel to it, for example, free-slip boundary conditions are applied. In order to strain the model, walls are moved inwards to apply compression or outwards to apply extension. Once a wall is moved all particles in the deformation box are moved assuming homogeneous deformation. Then the relaxation algorithm starts. Once the lattice is in an equilibrium condition after a deformation step, all springs are checked for breaking. If the stress associated with a spring reaches a critical tensile value, fracturing is induced by breaking springs (Fig. 3c). The spring with the highest probability of breaking will do so and a new relaxation routine will start. This procedure is repeated until all springs are below the breaking threshold and a new deformation step is applied. The fracturing process is therefore highly nonlinear; the whole lattice can fail at once between two deformation steps. This microscopic failure criterion will result in shear and extension failure between two neighbouring particles because particles are connected with six springs to their neighbours. Particles with broken springs are still repulsive. The coupling of the discrete code with the 'Elle' environment is performed as follows. In 'Elle' the microstructure of a rock is defined by nodes, which are connected with boundary
14
D. KOEHN E T A L .
Fig. 4. (a) Combination of the 'Elle' modelling environment with the discrete model. In Elle, grains are defined by double and triple nodes that are connected by straight segments. Particles of the discrete model that lie within an Elle grain have the same properties. (b) Example of a microstructurethat was used for most of the simulations shown in this paper. segments that define grains (Fig. 4a). In the discrete code grains are filled with particles. All particles within a grain can have specific parameters like the elastic constant and the breaking strength of springs. Springs that connect particles of different grains define grain boundaries and have different properties to springs that lie within grains. In all the presented models grain boundary springs have about half the tensile strength of intragrain springs. This value is chosen in order to have an influence of grain shapes on fracture development. Therefore most fractures will be intergranular. Intragranular fractures mainly appear when shear fractures develop. We expect grain boundaries to have a lower cohesion due to the presence of fluids or impurities. In order to reduce lattice effects, noise is distributed in the model by three different methods. First an initial microstructure is drawn and used as an input file with a certain spatial distribution of grain boundaries (Fig. 4b). Then a statistical distribution is applied to the elastic constants of different grains. Each grain has an individual elastic constant chosen by a random function from a gauss distribution. The mean elastic constant of the whole deformation box is represented by the mean value of the distribution. The breaking strength of all springs is statistically dispersed using a linear distribution because a rock is also expected to be disordered on a scale smaller than the grain size (Malthe-SOrenssen et al. 1998b). This distribution is not dependent on grains and grain boundary springs still have about half the breaking strength of intragrain springs.
The presented simulations have a resolution of 184 800 particles within the deformation box. In all simulations we strain the model in small steps of 0.02%. In the model the unscaled mean value of the initial elastic constant is 1.0 and the breaking strength of intragranular springs is 0.006. Grain boundaries always have half the breaking strength of intragranular springs. We scale these values in the results section in order to discuss the models in a geological context. We assume that the breaking strength of grain boundaries is 30MPa. Therefore a value of 0.0001 in the model scales to 1 MPa. An initial spring constant of 1.0 thus represents an elastic constant of 10 GPa. This leads to a mean tensile strength of the aggregates used in the simulations of about 25 MPa and a mean compressive strength of about 90 MPa under pure shear deformation (starting from zero strain and zero stress). This compressive strength is relatively low because in pure shear deformation one principal stress is tensile if the experiment is not preloaded. If this stress is also compressive the breaking strength will be considerably higher. This relation is also seen in laboratory experiments and is predicted by theory (Jaeger & Cook 1976; Paterson 1978). If the aggregate in our simulations is preloaded before tectonic deformation is applied (the confining pressure is increased), the compressive strength rises to about 140-190 MPa.
Results We present a number of simulations of fracture development with different deformation histories
FRACTURE AND VEIN PATTERNS and show the dependence of the pattern on the initial noise in the system and differences in the mean elastic constant of the aggregate. First we discuss fracture patterns that develop under pure shear deformation. We investigate the influence of noise and mean elastic constant of the aggregate on pattern formation using examples that experience the same pure shear deformation history. Then we move on to discuss deformation patterns that develop during area increase (expanding the model area). Finally we include complex deformation histories where samples experience an initial uniaxial loading and then a tectonic deformation. This leads to a number of deformation histories, which produce the vein and fracture patterns geologists can find in some field areas.
mean stress and the differential stress. Note that compressive stress is negative. Figure 5b shows schematic Mohr circles that represent stress states for some of the deformation events shown in Figure 5a. The developing patterns are illustrated in Figure 5c. Starting from a completely relaxed state, during progressive deformation the stress field behaves according to linear elasticity theory. Principal stresses are oriented parallel to the directions of compression and extension and increase linearly with the same slope into the tensile and compressive field. Consequently the mean stress remains zero and the differential stress curve has a slope that is twice as steep as that of a single principal stress component (Fig. 5a). The corresponding Mohr circle grows but remains stationary at the intersection of the axis of the diagram. A peak in the curve of the tensile stress marks the intersection when the Mohr circle reaches the tensile part of the failure envelope. At this point tensile fractures grow in the sample perpendicular to the extension direction. They cross most of the sample and show a distinct spacing. These fractures accommodate extensional strain and relax tensile stresses. This effect is also seen in the stress-strain curve in Figure 5a. The tensile
Pure shear deformation First we examine fracture patterns that are generated in a sample that experiences pure shear deformation. The initial sample is not stressed, deformation starts from a completely relaxed state. Figure 5a shows the stress-strain relationship during deformation. Different curves correspond to the maximum and minimum stress, the
~(uPa) (a)
400
A(~(~~m
8o
-40
I
0
"
-
0.01
0.0Z
E
15
(b)~l;~n ~
(+)
~
/
(')
A~ T
,~
Fig. 5. Simulation of pure shear deformation of an aggregate. (a) Stress-strain relation showing the two principal stresses (oh, ~r2),the mean stress (Crm)and the differentialstress (Ao-).(b) SchematicMohr circle diagram of (a) showing Mohr circles during different stages of the simulation. They mark the initial starting configuration,mode I failure and mode II failure. Note that the x-axis is negative towards the right-hand side since compressivestresses are negative in this paper. (c) Three successive stages during the simulation. The picture on the left-hand side shows development of mode I fractures. The pictures in the middle and on the right-hand side show development of mode II shear fractures.
16
D. KOEHN ET AL.
stress drops, the mean stress starts to increase and deviates from zero and the differential stress has a slope that is less steep. The compressive stress is not much affected by extension fractures because the vertical load is still supported by the sample. The Mohr circle in Figure 5b starts to move from the tensile regime into the compressive region and continues to grow in radius. The next failure event is reached when the differential stress is high enough so that the Mohr circle cuts the upper part of the failure curve and shear fractures develop. Before this moment is reached the stress curves show increasingly flattening slopes due to the local development of small shear fractures prior to failure of the whole sample. The failure of the sample is accommodated by slip along shear fractures (shear fracture becomes a distinct fault), which relaxes the compressive stress in the sample in a number of steps as seen in the stress-strain curve. The differential stress is also relaxed until the sample reaches a quasi steady-state where stresses remain almost constant and the bulk behaviour is almost purely plastic. This deformation history will result in the primary growth of extension fractures and a secondary growth of conjugate shear fractures.
(a}
t
Extension fractures will progressively open to accommodate extensional strain so that veins can form whereas the shear fractures will mainly show slip along their surfaces and develop into faults. Thus, in a natural example one would expect only one set of veins that develops out of extension fractures.
Differences in initial noise and elastic constant Differences in initial statistical noise and in the mean elastic constant of the aggregate influence the developing pattern. By initial noise in the system we mean the width of the distribution of elastic constants of different grains and the breaking strength of individual grains. Figure 6 shows two simulations with identical initial microstructure and elastic constants as in Figure 5, but with varying width of the distribution of the breaking strengths. The simulation shown in Figure 6a has a distribution of breaking strengths from 24 to 36 MPa whereas the simulation shown in Figure 6b has a distribution of breaking strengths ranging from 12 to 48 MPa. For each simulation two successive stages at the same amount of strain are shown, deformation is
) i
,.
...,~ :
~,
/
/ ( ?
(b)
./
3,
9 4
'!
.
Fig. 6. Two simulations with the same boundary conditions but different initial distributions of breaking thresholds, where (a) has a narrower distribution of breaking thresholds than (b). (a) Less noise in the system produces stronger localization of structures. (b) More noise produces more dispersed fractures with different sizes.
FRACTURE AND VEIN PATTERNS pure shear where compression is vertical. Stage one shows the initial development of extension fractures and stage two the development of secondary shear fractures. Two important differences can be observed between the different distributions of breaking strengths. First a wider distribution of breaking strengths results in a more dispersed development of fractures whereas a narrow distribution results in localized fractures that cut the whole aggregate. Secondly, the absolute breaking strength is lowered using a wider distribution of breaking strengths; more fractures develop in Figure 6b than in Figure 6a. The first effect can be explained as follows. A narrow distribution of breaking strength, which means that the initial noise in the system is low, represents a very brittle material. A lot of springs in the aggregate will reach their tensile strength within a single deformation step. This will result in an almost instantaneous propagation of large fractures through the aggregate. In particular, an extension fracture will continue to propagate once it nucleates since tensile stresses at the fracture tip will increase while the fracture grows in length. Once a large fracture is present it will be able to relax stresses
(a)
in the surrounding aggregate so that additional fractures will only develop at a certain distance to the initial fracture. This stress shielding mechanism produces a distinct spacing of fractures in the aggregate. If the width of the distribution of breaking strengths in the model is larger (Fig. 6b), fractures can develop at lower strains. This will reduce the overall breaking strength of the aggregate. Since the strength of springs varies significantly, fractures that nucleate may stop propagating. Therefore a range of fractures develops initially with different lengths. More fractures nucleate in an aggregate with a wider distribution of breaking strengths and the developing spacing of fractures will not be as distinct as in the example with the narrow distribution. The material has a bulk response to deformation that is less brittle. The effect of the mean elastic constant (Young's modulus) of the aggregate on pattern formation and rheological behaviour is shown in Figure 7. The mean Young's modulus of the aggregate increases from Figure 7a to e. Figure 7a to e shows an aggregate with a mean Young's modulus of 1, 2.5, 5, 7.5, and 12.5GPa. All simulations were performed
(b)
,( '. 2
"
17
Cc)
" '
it
!
!
0
i
9 ,
~
_,oi ";'F""
/ I
I 0 . . . . . . . . . . . . . . 0.01 . . . . . .
-~oI 0
~0,Ol
0,0~ E
mode,,
I~
0.02
Fig. 7. (a) to (e) show simulations under the same pure shear conditions with increasing mean elastic constant. A lower elastic constant produces stronger localization and a wider spacing of mode I fractures. If the elastic constant is higher, mode II fractures dominate. (f) Stress-strain relation for (a). Mode I failure shows a sawtooth curve typical for fast propagation of large mode I fractures. (g) Stress-strain relation for (e). Mode I failure is continuous with no sudden drop of the tensile stress (02 curve). Mode II failure results in a drop of the differential stress.
D. K O E H N ET AL.
18
using pure shear deformation and the snapshots presented are taken at similar amounts of finite strain. The breaking strength of springs in all simulations has the same distribution and mean value. Consequently failure will occur at lower strains in an aggregate that has a higher elastic constant. Therefore the number of fractures increases from Figure 7a to e. In addition, the compressive stresses are not high enough to induce shear failure in Figure 7a and only to a minor extent in Figure 7b and c, whereas shear fractures dominate in Figure 7e. An aggregate with a lower elastic constant localizes extensional strain in tensile fractures, which open and form large veins in Figure 7a and to a minor extent in Figure 7b and c. Figure 7d shows a combination of extension fractures and shear fractures. The spacing of extension fractures depends on the mean elastic constant of the aggregate. In the simulation in Figure 7a one large vein develops within the model whereas Figure 7b shows two and Figure 7c three large veins. The spacing is reduced with an increase in Young's modulus. A softer material with a lower elastic constant can relax a larger region around an opening fracture and thus produces a larger spacing than a material with a higher Young's modulus. This relation becomes disturbed in Figure 7d and e, where shear fractures start to dominate the pattern. Spacing of shear fractures does not show a simple relation with increasing Young's modulus. Figure 7f shows the stress-strain relationship for the simulation presented in Figure 7a, and Figure 7g the stress-strain curve for the simulation presented in Figure 7e. These curves illustrate that the rheological behaviour of the aggregate also changes significantly with increasing elastic modulus. Failure due to the development of extension fractures shows significant differences in the stress-strain
(a)
't'
,.)
q
(b)
I n c r e a s e in a r e a
Figure 8 shows three simulations with an area increase of the deformation box. The simulations have horizontal and vertical tensile boundary conditions where the horizontal component is larger in Figure 8a and b. Figure 8a shows a simulation with a large differential stress where the horizontal stress component is three times as large as the vertical component. Figure 8b shows a simulation where the vertical stress component is about 70% of the horizontal stress component. Both extension components in the vertical and horizontal direction are identical in the simulation shown in Figure 8c. The fracture pattern in Figure 8a represents the dominant horizontal extension. A small number of horizontal fractures develop due to the vertical extension component. Figure 8b shows a different pattern where the dominance of the horizontal extension component can still be seen but the
(c)
.4
.r l
,f
relationship of the two examples and failure due to the development of shear fractures only occurs at the finite strain shown in the example of Figure 7g. Figure 7f shows a sudden drop of the tensile stress (0"2) at the beginning of extension fracture development. This indicates that the fractures develop relatively fast and grow large enough to invoke failure of the whole aggregate and reduce the tensile stress almost completely. The overall failure also reduces the differential stress. Figure 7g shows the aggregate with the highest Young's modulus in the presented sequence. It experiences a more continuous development of extension fractures that grow progressively while the aggregate reaches higher extensional strains. Therefore the tensile stress is released gradually and the differential stress continues to increase after extension fracture development.
i
9
,-r.
'~
"~
,~
~
..'
,'3-
(.
r
"1 (
>i
"],
r , N .tl
-is
. ,," ,J
"'t"
", J !_~
~
......... ]
Fig. 8. Three simulations with two extension directions and an increase in modelling area 9(a) Dominant horizontal extension produces well-oriented mode I fractures. (b) Both extension components become more similar so that fractures start to curve 9Orientation of fractures becomes more diffuse than in (a). (c) Horizontal and vertical extension components are the same, no relation of the orientation of fractures and the model boundaries can be observed. Fractures form polygons similar to mud cracks.
FRACTURE AND VEIN PATTERNS vertical extension component also influences fracture development. Fractures start to curve and lose a dominant orientation with respect to a principal stress direction because the stresses are similar. Fractures still produce a distinct spacing. In Figure 8c the stress field has no dominant extension direction and the developing fracture pattern shows no preferred orientation with respect to the simulation boundaries. Fracture development and orientation is controlled by the initial noise in the system. Once the first fractures develop they influence the stress field in the aggregate, which leads to a polygonization with a distinct spacing similar to the structures found in mud cracks in shrinking sediments (Suppe 1985).
Gravitational loading and pure shear deformation In a normal tectonic environment one would expect that sedimentary sequences are loaded during their deposition and then experience
,ool
19
tectonic strain afterwards (Price & Cosgrove 1990). In order to investigate the effects of initial gravitational loading on fracture and vein patterns several simulations were performed with complex deformation histories including different tectonic settings 9 In the first simulation with a two-stage deformation history we load the sample uniaxially and then apply a pure shear boundary condition. The sample is loaded vertically with fixed sidewalls. The vertical stress is proportional to the deformation steps and the horizontal stress is compressive due to Poisson effects. After a given amount of vertical strain we apply a pure shear deformation with vertical constrictional strain and horizontal extensional strain. The developing fracture patterns are shown in Figure 9a in three successive steps. Figure 9b shows the stress-strain diagram for the simulation and Figure 9c the corresponding Mohr diagram with four different Mohr circles a to 6 that are indicated in the stress-strain diagram of Figure 9b. The region on the left-hand side
(c)
(b)
't
T ~5
8
(+)
(+)
-t00 0
o0.005
0.01
Fig. 9. Complexdeformation history involving a compaction event followed by pure shear deformation. (a) Three successive fracture patterns during the simulation. Conjugate mode II fractures dominate the pattern. Only at the latest stage do small mode I fractures develop in the model in order to accommodate local tensile strain. (b) Stress-strain relation of the simulation. Both principal stresses are compressive during the compaction (regime L). Once tectonic deformation starts (regime T) the horizontal stress is relaxed but the differential stress increases. (c) Mohr circle diagrams for (b). The diagram on the left-hand side marks three successive stages during the compaction process. The diagram on the right-hand side shows the successive growth of the Mohr circle during tectonic deformation and the following failure that produces the dominant conjugate mode lI fractures. Note that the x-axis is negative towards the right-hand side since compressive stresses are negative in this paper.
20
D. KOEHN ET AL.
of the vertical line in Figure 9b indicates the gravitational loading regime (marked L for loading). The two principal stresses are compressive and the mean stress and differential stress increase successively. The development of the Mohr circle during initial loading is indicated in Figure 9c on the left-hand side where c~ marks the initial starting point and/3 and y the successive growth and movement of the Mohr circle into the compressive regime. After gravitational loading pure shear deformation shows an increase in compressive stress and decrease in tensile stress (Figure 9b in tectonic regime marked T). The mean stress stays constant and the differential stress increases with a steeper slope than in the loading regime. At point 6 in Figure 9b failure is initiated so that in Figure 9c the Mohr circle cuts the failure envelope. Since differential stress and mean stress are high the developing fractures are shear fractures with typical conjugate sets, which can be seen in Figure 9a. The Mohr circle will not reach the tensile part of the failure curve without crossing the failure curve in the compressive regime if pure shear deformation is applied after the uniaxial loading. The developing pattern will mainly consist of conjugate shear fractures. Small extension fractures will only grow at a larger amount of deformation in order to accommodate local tensile strains as can be seen in Figure 9a. However, if fluid pressure is involved, the Mohr circle may cross the failure curve in the tensile regime during deformation (see discussion). Gravitational loading and area increase
In this section we take a look at fracture patterns that develop during different kinds of initial loading followed by a tectonic deformation that is characterized by a large area increase so that both major components of the strain in the modelled plane are extensional. The developing fracture patterns and corresponding stress-strain curves are shown in Figure 10a to d. Figure 10a shows a simulation with a small amount of initial loading. Tectonic deformation is dominant with a large horizontal extension component. During extension the compressive stresses are relaxed and the horizontal stress component becomes tensile, which results in the development of vertical extension fractures that are opening. Failure of the whole aggregate is fast, which is illustrated by the sawtooth stress curves in Figure 10a. The material behaves in a brittle manner and shows a strong localization of strain in two to three tensile fractures that are opening and that show a distinct spacing.
Following failure the aggregate reaches a quasi steady-state plastic behaviour where stresses remain at relatively constant values while the strain increases. Small fluctuations in the tensile stress represent growth of secondary extension fractures to accommodate strain. Figure 10b shows a simulation where gravitational loading is dominating. Failure of the aggregate starts during initial loading when a peak differential stress is reached. During this stage conjugate sets of shear fractures develop. Successive tectonic deformation reduces all stress components and leads to a local growth of small extension fractures. The resulting fracture pattern is very similar to that of Figure 9, where a smaller amount of initial loading was followed by a pure shear deformation. Both simulations show a dominant primary growth of conjugate shear fractures and a successive development of secondary small-scale extension fractures. An intermediate pattern between the simulations shown in Figure 10a and b is shown in Figure 10c. Here the aggregate is initially loaded by an intermediate amount so that initial loading itself does not induce failure. The aggregate is then extended in the horizontal and vertical direction where the horizontal extension component is dominant. During failure of the aggregate, conjugate shear fractures develop with a smaller angle towards the principal compressive stress direction than in Figure 10b. The fractures in Figure 10c are of the hybrid extension/shear type of Price & Cosgrove (1990), an intermediate type of fracture. In the simulation shown in Figure 10c these conjugate fractures accommodate successive amounts of extensional strain and are opening. Figure 10d shows a different type of gravitational loading where the aggregate is initially loaded by the same horizontal and vertical amount. Therefore the differential stress stays at zero and the two principal stress components as well as the mean stress increase by the same amount. The following tectonic deformation has a horizontal extension component. During extension all stresses are relaxed. The differential stress starts to increase as a result of the large horizontal extension component. This resulted in the growth of conjugate hybrid extension/ shear fractures similar to those shown in the simulation of Figure 10c. In summary gravitational loading can lead to the development of primary conjugate sets of hybrid extension or shear fractures that will accommodate extensional strain during successive deformation. It can be seen therefore that fractures and veins in a rock have a record of
FRACTURE AND VEIN PATTERNS
(a)
'LJ;............. ,.i' ;
}
,
(b)
,. '...
,
0
(MPa L
21
r
100 o (
o
o
~o21
AO
zc
O
/
o~
o~
i
lOO
-20
0
0.02
0.04
-20
13
01
0
0.01
E
-40
-40 0
0.01
0.0Z s
0
O,OOS
s
Fig. 10. Four different simulations and the corresponding stress-strain curves for deformation histories involving different kinds of compaction that are followed by tectonic deformation that involves extension in the modelling plane. (a) Small amount of compaction results in a situation where mode I fractures are dominant. (b) Large amount of compaction results in the development of conjugate mode II fractures during compaction. The following tectonic extension produces local mode I fractures but strain is mainly accommodated by the already existing fracture network. (c) A simulation with intermediate compaction followed by tectonic extension results in the growth of hybrid extension/shear fractures that are opening. (d) Compaction that is hydrostatic can also lead to hybrid extension/shear fractures if the following tectonic deformation is strongly non-hydrostatic so that extension is dominating in the horizontal direction.
22
D. KOEHN E T A L .
the gravitational loading of the rock prior to tectonic deformation.
Discussion We compare the results of the numerical study with the vein sets shown in Figure 2. The vein in Figure 2a may have started as a hybrid extension fracture and may reflect a stress state that would produce that type of fracture. This stress state would then be represented by the whole history of the vein development. However, our simulations show that quite a number of different stress states may develop during a deformation history. Therefore this vein could also form as a shear fracture (or hybrid shear depending on its orientation relative to the principal stress axis) initially and then open afterwards due to an extension component during deformation. Such a setting could also produce vein patterns similar to the ones shown in Figure 2b where conjugate shear (or hybrid shear) fractures may form first and then later extension takes over and the initial fractures are opening as veins. However, further study on the internal geometries of such vein sets is needed in order to prove that the proposed scenario can develop in nature. We use hyperbolic Mohr circle diagrams to illustrate the stress states and failure in our simulations. There is still much debate on whether a hyperbolic Mohr circle can be used for intermediate failure types between mode I and mode II, namely hybrid extension/shear fractures. Engelder (1999) illustrates that these intermediate structures cannot be explained as propagating cracks by linear elastic fracture mechanics theory. He rather concludes that they may form by an out-of-plane propagation of mode I fractures that are subject to a shear stress. In our simulation a number of scenarios produce patterns that are dominated either by mode I or mode II fractures. In intermediate settings quite often a combination of mode I and mode II fractures develop where both failure modes still show distinct sets of fractures. Only in a narrow range do we find fractures that have an intermediate orientation (between mode I and II), which fall into the category of hybrid extension/shear fractures (Fig. 10c). These may well be combinations of mode I and mode II fractures as proposed by Engelder (1999). It is interesting to see that in a number of simulations quasi steady-state behaviours are reached (Figs 5a and 10a, c). The fracture systems that develop can accommodate additional strain with only minor increase in stress, which is
probably due to additional growth of small fractures or friction along shear fractures. Prior to this quasi steady-state behaviour, stresses drop significantly when the whole systems fails. The remaining tensile stress component is very small since most tensile stresses can be accommodated by opening fractures or veins. Compressive stresses are higher and can probably be related to friction along shear fractures as long as one component of the stress tensor in the modelling plane is compressive. As long as the system does not heal after failure stresses will probably remain low. If this steady state is also reached in natural rocks, it has a strong influence on the strength of brittle faults and earthquake behaviour (Scholz 2002). Once a rock fails by fracturing it may deform in a plastic way without significant increase in stress (Paterson 1978; Zhang et al. 1990; Karato 1995). Once a fault forms and develops a cataclasite it may behave as a plastic material with no major increase in stress if fractures are not healing. Therefore no successive earthquakes will be expected along such a fault. However, if veins grow and fill the fractures, the cataclasite may heal and retain a certain strength again. Then it may fracture by catastrophic failure at high stresses (Scholz 2002). What we have not yet taken into account is the effect of fluid pressure on the developing fracture patterns. This is certainly an important component that we will include in the future. A high fluid pressure can shift the Mohr circle towards the tensile regime so that mode I fracture may develop during the loading history instead of conjugate shear fractures (Price & Cosgrove 1990; FlekkCy et al. 2002). However, conjugate shear fractures may also develop in this case if the differential stress is already high due to a non-hydrostatic initial loading. Fluid pressure and developing fractures will influence each other depending on how effectively fractures can drain an existing fluid pressure. If the tectonic stress is non-hydrostatic or if fluid pressure gradients exist along anisotropies, fluid pressure will provoke failure (Bercovici et al. 2001). Once fractures develop, the fluid pressure may drop because the permeability of the rock increases. Whether or not the fluid pressure influenced fracture and vein development in the natural samples that were presented in this paper is not clear. Bedding parallel veins in the area may have formed during high fluid pressures because they are located along the bedding planes between porous sandstones and less porous mudstones. Fluid pressure gradients can be expected along these anisotropies, which may invoke failure. The veins that are presented
FRACTURE AND VEIN PATTERNS in this paper are however located within the more porous sandstones and are probably more associated with tensile strain than a higher fluid pressure. However, additional research is needed in order to fully understand the effect of fluid pressure on the development of these systems. In the following paragraphs we will discuss some additional boundary effects of the model that we used. The box boundaries have an influence on the development of shear fractures. Box boundaries reflect shear fractures because the boundaries behave like rigid pistons and boundary particles are fixed perpendicular to the walls. They cannot accommodate a shear displacement of the boundary so that shear fractures that hit the walls are repelled and a new shear fracture grows back into the model. This effect will influence the density of shear fractures in the model and also their spacing. In some cases shear fractures will dominate near to the compressive walls (upper and lower wall) whereas extension fractures dominate in the centre of the model (Fig. 7). However, this effect is not always observed (Fig. 10c). An additional problem arises because the underlying lattice is triangular. Therefore fracturing is partly anisotropic, which is especially important in the case of shear fractures where slip should take place. The anisotropies of the lattice can be overcome by adding distributions to the elastic constants and spring breaking strengths that are larger than the lattice anisotropies (Malthe-SCrenssen et al. 1998b). Mode I fractures show no lattice-preferred orientation in our model, which can be best observed in Figure 8c where stresses are hydrostatic and no lattice directions nor any effects of the model boundaries are observed. Shear fractures may however still be influenced by the lattice (Fig. 10b) but can also develop in non-lattice directions (Fig. 10c). One has to note that the direction of the shear fractures in Figure 10b is more or less a lattice direction, but also the direction where the fractures would develop anyhow. The threshold of relaxation in the model can also change the developing pattern. This is recorded in detail in Malthe-SCrenssen et al. (1998b). Since we are moving all particles assuming homogeneous strain in the model before relaxation starts and since our relaxation threshold is very small relative to the applied strain, it has no major influence on the developing pattern. However, if a significantly higher relaxation threshold is used, different spacing of mode I fractures will result since the system will be progressively slower and relaxation will be more localized.
23
Conclusions Modelling of progressive fracture development in aggregates shows that different type of fractures can grow during one tectonic deformation event. Pure shear deformation produces primary extension fractures that are followed by a secondary set of conjugate shear fractures. These patterns change depending on the properties of the aggregate, namely its mean elastic constant and the initial noise in the system. Pure extension of the modelling area leads to one dominant extension fracture set or to a polygonal set of fractures if the principal stresses are equal. A more complex deformation history that includes gravitational loading as well as tectonic strain can lead to the development of conjugate shear fracture sets that can accommodate extensional strain and thus form veins during later stages of deformation. These conjugate sets of veins may be followed by secondary extension fractures. These complex veins may record gravitational loading preceding the tectonic deformation. Their orientation can be used to find the orientation of the non-hydrostatic stress field that existed during gravitational loading or tectonic deformation. We thank T. Engelder and J. Cosgrove for constructive reviews. We also thank A. Malthe-SCrenssen for his help with the discrete element code. JA acknowledges funding by the DFG-Graduiertenkolleg 'Composition and Evolution of Crust and Mantle'.
References ALLEN, D. M. DE G. 1954. Relaxation Methods. McGraw Hill, New York. ANDERSON,E. M. 1951. The Dynamics of Faulting and Dike Formation. Oliver and Boyd, Edinburgh. AYDIN, A. & RECHES, Z. 1982. Number and orientation of fault sets in the field and in experiments. Geology, 10, 107-112. BERCOVICI, D., RICARD, Y. & SCHUBERT, G. 2001. A two phase model for compaction and damage, 3: Applications to shear localization and plate boundary formation. Journal of Geophysical Research, 106, 8925-8942. BONNET, E., BOUR, O., ODLING, N. E., DAVY, P., MAIN, I., COWIE, P. & BERKOWITZ, B. 2001. Scaling of fracture systems in geological media. Reviews in Geophysics, 39, 347-383. ENGELDER, T. 1999. The transitional-tensile fracture: A status report. Journal of Structural Geology, 21, 1049-1055. FLEKK~3Y, E. G., MALTHE-S~3RENSSEN, A. & JAMTVEIT, B. 2002. Modeling hydrofracture. Journal of Geophysical Research B8, ECV 1, 1- 11. GRI~ITH, A. A. 1920. The phenomena of rupture and flow in solids. Transactions of the Royal Socie~ Series A, 221, 163-198.
24
D. KOEHN ET AL.
HANCOCK, P. L. 1985. Brittle microtectonics: principles and practice. Journal of Structural Geology, 7, 437-457. HUDSON, J. A. 8z COSGROVE,J. 1997. Integrated structural geology and engineering rock mechanics approach to site characterization. International Journal of Rock Mechanics and Mining Science, 34, 577. JAEGER, J. C. 8z COOK, N. G. W. 1976. Fundamentals of Rock Mechanics. Chapman and Hall, London. JAMTVEIT, B., AUSTRHEIM, H. 8z MALTHESORENSSEN, A. 2000. Accelerated hydration of the Earth's deep crust induced by stress perturbations. Nature, 408, 75-78. JESSELL, M. W., BONS, P. D., EVANS, L., BARR, T. D. & STt3WE, K. 2001. Elle, The numerical simulation of metamorphic and deformation textures. Computers and Geosciences, 27, 17-30. KARATO, S. 1995. Rock deformation: Ductile and brittle. Reviews in Geophysics, American Geophysical Union, 33. KRANTZ, R. W. 1988. Multiple fault sets and threedimensional strain; theory and application. Journal of Structural Geology, 10, 225-237. MALTHE-SORENSSEN,m., WALMANN,T., JAMTVEIT,B., FEDER, J. • J~SSANG, T. 1998a. Modeling and characterization of fracture patterns in the Vatnajokull glacier. Geology, 26, 931-934. MALTHE-SORENSSEN, A., WALMANN, T., FEDER, J., JOSSANG, T. & MEAKIN, P. 1998b. Simulation of extensional clay fractures. Physical Review E, 58(5), 5548-5564. MALTHE-SORENSSEN,A., WALMANN,T., JAMTVEIT,B., FEDER, J. 8z JOSSANG, T. 1999. Simulation and characterization of fracture patterns in glaciers. Journal of Geophysical Research, B104, 23157-23174. MEANS, W. D. 1976. Stress and Strain: Basic Concepts of Continuum Mechanics for Geologists. SpringerVerlag, New York.
OERTEL, G. 1965. The mechanism of faulting in clay experiments. Tectonophysics, 2, 343-393. OLIVER, N. H. S. & BONS, P. D. 2001. Mechanisms of fluid flow and fluid-rock interaction in fossil metamorphic hydrothermal systems inferred from veinwallrock patterns, geometry and microstructure. Geofluids, 1(2), 137. PATERSON, M. S. 1978. Experimental Rock Deformation - Brittle Field. Springer Verlag, Berlin. POLLARD, D. D. & SEGALL, P. 1987. Theoretical displacements and stresses near fractures in rocks: with applications to faults, joints, dikes and solution surfaces. In: ATKINSON, B. K. (ed) Fracture Mechanics of Rock. Academic Press, London, 277-348. PRICE, N. J. & COSGROVE, J. W. 1990. Analysis of Geological Structures. Cambridge University Press, Cambridge. RAMSAY, M. R. & CHESTER, F. M. 2004. Hybrid fracture and the transition from extension fracture to shear fracture. Nature, 428, 63-66. RAMSAY, J. G. & HUBER, M. I. 1983. The Techniques of Modern Structural Geology, 1: Strain Analysis. Academic Press, London. RANALLI, G. 1995. Rheology of the Earth, 2nd edn. Chapman & Hall, London, UK. SCHOLZ, C. H. 2002. The Mechanics of Earthquakes and Faulting, 2nd edn. Cambridge University Press, Cambridge. SUPPE, J. 1985. Principles of Structural Geology. Prentice-Hall, New Jersey. WALMANN, T., MALTHE-SORENSSEN, A., FEDER, J., JOSSANG, T., MEAKIN, P. ~r HARDY, H. H. 1996. Scaling relations for the lengths and widths of fractures. Physical Review Letters, 77, 5292-5296. ZHANG, J., WONG, T.-F. & DAVIS, D. M. 1990. Micromechanics of pressure-induced grain crushing in porous rocks. Journal of Geophysical Research, 95, 341-352.
Segmentation and interaction of normal faults within the Colfiorito fault system (central Italy) F. M I R A B E L L A , V. B O C C A L I & M. R. B A R C H I Geologia strutturale e Geofisica, Dipartimento di Scienze della Terra, Universith di Perugia, piazza Universith 1, 06100 Perugia, Italy (e-mail: mirabell@ unipg, it) Abstract: Fault segments belonging to a fault population can link and interact, eventually
forming a single larger fault, and thus affecting the estimation of the maximum expected earthquake. We present throw distribution data along the Quaternary normal faults of the Colfiorito fault system (central Italy), which consists of four main fault segments and where a seismic sequence occurred in 1997-1998. Throw values along the two central overlapping, en-rchelon segments (8.5-9.5 km long) were measured on a good stratigraphic marker, by constructing a set of closely spaced geological cross-sections, perpendicular to the fault strike. As these faults are commonly retained active and border Quaternary basins, we compare morphological and geological throws in order to verify the faults neotectonic activity. Geological and morphological throw distributions show good correlation, testifying that recent faulting affected the topographic surface and suggesting that the observed offset completely accumulated during the Quaternary. The throw distribution along the fault segments is asymmetric and reaches maximum values (500-550m) within the zone of fault overlap, suggesting mechanical interaction between the studied faults. Maximum length-throw correlation suggests that the studied faults grew according to a linear scaling relationship.
Faults are usually organized in systems of segments, which may interact and link forming a single fault (e.g. Anders & Schlische 1994; Cartwright et al. 1995). Fault interaction and geometrical irregularities between individual faults play an important role in the initiation and termination of seismic ruptures and in the distribution of aftershocks (e.g. Aki 1979; King & Nabelek 1985; Sibson 1986; Barka & Kadinsky-Cade 1988; Sieh et al. 1993; Amato et al. 1998; Chiaraluce et al. 2003). Fault discontinuities and interaction are also important in constraining fault growth conceptual models that consider either single continuous fault planes that increase their displacement as they grow (Watterson 1986; Walsh & Watterson 1988; Marrett & Allmendinger 1991; Cowie & Scholz 1992) or fault growth by segment linkage (Segall & Pollard 1980; Burgmann et al. 1994; Willemse et al. 1996). The primary expression of interaction between fault segments is on the way they grow and is reflected by the shape of the displacement profiles that show asymmetric slip distribution along strike (Peacock 1991; Gupta & Scholz 2000). The irregular way in which faults grow has also been attributed to the presence of
heterogeneities in shear strength but this effect is supposed to be of less importance than that due to fault interaction (Peacock & Sanderson 1991). This has also been suggested by numerical modelling, which indicates that mechanical interaction between neighbouring faults, even if not connected, can cause asymmetric slip distribution (Willemse et al. 1996). As fault dimensions are important for seismic hazard purposes (Wells & Coppersmith 1994), the interaction between adjacent faults affects the estimation of the maximum possible earthquake. Hence the possibility that two or more fault segments can interact to form a larger fault becomes a matter of interest, especially when studying active fault systems in populated areas. This paper is dedicated to the Colfiorito Fault System (CoFS) in the Umbria-Marche region (central Italy) where a seismic sequence characterized by a series of moderate magnitude earthquakes occurred in 1997-1998. We obtained the fault throw distribution along the two central segments of the fault system, where recent surface mapping (Barchi et al. 2001) provided a detailed fault pattern. From surface data, a set of 500 m spaced geological cross-sections oriented perpendicularly to fault strike was
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 25-36. 0305-8719/05/$15.00 ,:~ The Geological Society of London 2005.
26
F. MIRABELLA E T A L
constructed in order to draw the along-strike distribution. As recent fault activity is commonly supposed to affect the topographic surface (e.g. Burbank & Anderson 2001), we compare the vertical component of displacement (geological throw) with the difference between the maximum and minimum topographic elevations (morphological throw), measured at the footwall and hangingwall respectively, in order to discuss the neotectonic fault activity. Finally, by plotting maximum throw versus length for most of the normal faults cropping out in the investigated area, we derive a scaling law for the fault growth.
Geological setting The Colfiorito Fault System belongs to the active alignment of normal faults (Umbria Fault System - UFS, Barchi 2002) of the UmbriaMarche Apennines (Fig. la). The UFS faults border the most important N W - S E trending Quaternary continental basins (Gubbio, Colfiorito, Norcia, Cascia and Castelluccio basins).
The faults of the UFS are commonly referred to as active faults, on the basis of geomorphological (e.g. Ficcarelli & Mazza 1990; Coltorti et al. 1998; Messina et al. 1999), geological (e.g. Lavecchia et al. 1994; Calamita et al. 1999; Barchi et al. 2000; Boncio & Lavecchia 2000; Mirabella & Pucci 2002) and seismological evidence (Deschamps et al. 1984; Haessler et al. 1988; Amato et al. 1998; Ekstroem et al. 1998; Mariucci et al. 1999; Stramondo et al. 1999; Barba & Basili 2000). Historical seismicity (maximum intensity ---- 10 of the modified Mercalli scale, MercalliCancani-Sieberg intensity scale, MCS, CPTI 1999) and recent earthquakes (Norcia, 1979 Ms = 5.5; Gubbio, 1984 Ms = 5.3; Colfiorito, 1997-1998 Mw ~ 6.0; Deschamps et al. 1984, 2000; Haessler et al. 1988) indicate that the presently active stress field (extensional with S W - N E trending o-3) is consistent with the geological long-term stress field obtained by structural data and active since the Quaternary (Lavecchia et al. 1994; Mariucci et al. 1999). The Colfiorito Fault System is presently active. The area was recently struck by a long
Fig. 1. (a) Schematic structural map of the Umbria-Marche region showing the alignment of the intramontane basins along the Umbria Fault System (UFS). Historical seismicity is reported for the period 461 BC to AD 1979 (Boschi et al. 1997). Focal mechanisms and magnitudes are for the 1997-1998 Colfiorito sequence (Ekstroem et al. 1998), for the 1979 Norcia earthquake (Deschamps et al. 1984) and for the Gubbio earthquake (Dziewonski et al. 1985). (b) Schematic pattern of the Colfiorito Fault System and location of the four main segments.
SEGMENTATION AND INTERACTION OF NORMAL FAULTS 1997-1998 seismic sequence characterized by six main earthquakes with 5 < Mw < 6 all nucleated on SW-dipping normal faults (Chiaraluce et al. 2003). On 26 September 1997, two moderate magnitude main shocks (Mw = 5.8 and Mw = 6.0) occurred in a short time period (9 hours), rupturing in opposite directions and with hypocentres located at a distance of about 2 - 3 km from each other (Pino & Mazza 2000). At the surface, the pattern of the CoFS faults is complex. Several fault segments mainly strike about N135 ~ and dip to the SW, dips ranging from 50 ~ to 80 ~ Subordinate NS, EW and N E - S W normal faults are also present in the area. The faults dissect pre-existing compressional features related to the Umbria-Marche fold and thrust belt and some of them border the eastern side of the intramontane Pleistocene basins (Colle Croce-Annifo, Colfiorito, Cesi, S. Martino). In particular, the Colfiorito basin consists of lower-middle Pleistocene fluvio-lacustrine deposits (Ficcarelli & Mazza 1990; Coltorti et al. 1998) that reach a thickness of about 120 m in the depocentre (Messina et al. 1999). Despite the complexities of the fault arrangement, with patterns of overstepping synthetic and antithetic splays at different scales of observation, four main fault segments with greatest continuity are recognized, from North to South: M. Pennino, M. LeScalette, Cesi-S. Martino and M. Cavallo-M. Fema (Fig. lb). These main segments have similar characteristics, that is, they strike N W - S E , have similar along-strike length (ranging from 8 to 10 kin) and similar maximum throws of up to 550-600 m (Barchi et al. 2000; Mirabella & Pucci 2002). A structural sketch (Fig. 2a) of the two central segments (M. LeScalette and Cesi-S. Martino) is drawn from the recent detailed geological map (1:10 000 scale) of Barchi et al. (2001). The M. LeScalette segment comprises a major fault and subparallel faults, strikes ranging from N130 ~ to N140 ~ (Mirabella & Pucci 2002). Several minor faults are also present, strikes varying between N120 ~ and about N160 ~ The most continuous fault plane borders the Colfiorito basin and is in the northwestern part of the segment. It puts in contact debris deposits in the hangingwall with Cretaceous limestone (Maiolica fro, tithonian-aptian) in the footwall. Along this fault, 2 to 20 cm high free faces were locally exposed and interpreted by some authors as due to coseismic reactivation during the 1997 earthquake (Cello et al. 1998; Calamita et al. 1999). The Cesi-S. Martino segment, about 8.5 km long, is characterized by a greater variability in
27
the fault attitude (Fig. 2a). From NW to SE, it comprises the following three portions: the first that borders the Cesi basin, strikes about N130 ~ with dips in the range 55-60~ the second that borders the S. Martino basin, strikes about NI70 ~ with dips of 62-74~ the third strikes N150 ~ and dips about 60 ~ The Cesi-S. Martino segment is shifted about 3 km to the SW with respect to the M. LeScalette segment (Fig. 2b). The overlap zone is characterized by a complex pattern of SW dipping and subordinate NE dipping normal faults with strikes ranging from NI20 to N170 ~ On the basis of striated fault planes, the normal faults are prevalently dip-slip, although some planes possess a strike-slip component (Fig. 2c). The inversion of fault slip data, although lacking a clear antithetic system, indicates a roughly vertical 0-1 and S W - N E trending 0-3 (Barba & Basili 200O).
Throw profiles along the Colfiorito Fault System The throw values of the normal faults in the study area were measured along a set of 27 geological cross-sections through the fault segments. The cross-sections, numbered from SI to $27 (Fig. 2a), are oriented N55 ~ with a mean spacing of about 500 m and are perpendicular to the average fault strike. The Cesi-S. Martino segment is characterized by the presence of a single fault ( N i l ) , about 8.2 km long, on which all the measured throw has accumulated, whereas for the M. LeScalette segment the total throw is distributed on six smaller faults (N2, N3, N6, N7, N8, N9 in Fig. 2a). Considering the dextral en-~chelon array of the normal faults (Fig. 2a), the group of sections S1 to S12 shows the geometry of the M. LeScalette segment, while the group of sections S12 to $24 displays the geometry of the Cesi-S. Martino segment. The overlapping area between the two major fault segments is imaged by section S12 (Fig. 2a). Sections $25, $26, and $27 have been drawn in order to measure the throw distribution of small faults in the southernmost region, which are thought to have been activated during the 1997-1998 earthquake (Basili & Meghraoui 2001). Three sections, representative of the structural style of the CoFS, are represented in Figure 3a. Section $9 (Fig. 3a) shows that the hangingwall of the M. LeScalette faults comprises gently folded Cretaceous limestones and marls involved in pre-existing compressional structures (thrusts, transpressive faults and associated folds). The
28
F. M I R A B E L L A ET AL.
Fig. 2. (a) Schematic structural map of the study area derived from 1 : 10 000 mapping scale with intramontane basins and mapped faults. The studied normal faults are numbered from 1 (N1) to 15 (NI5). Thin lines labelled from S1 to $27 show the location of the 27 geological cross-sections used for the measurment of geological and morphological throws. (b) Stereographic projections (Schmidt equal area projection, lower hemisphere) of fault slip data measured along the main faults' scarps and along minor faults in the study area (after Barba & Basili 2000). (c) Schematic pattern of the main fault segments studied in this work and parameters of length and separation.
SEGMENTATION AND INTERACTION OF NORMAL FAULTS
Stratigraphy SW Oligocene __qZ~------~ middle-Eocene LowerTuronian I I [ upperAlbian ~ ~/11 Aptian-Albian ~ ~r~ LowerAptian~---~----~ UpperTithonian 0 Middle Tithonia~ I Toarcian
29
M.LeScalettefaults N7 N3 | N8 NE
i
0
SW .
1
Cesi S.Martinofault ......... " a~ ' %
M.LeScalettefaults
Nil
-NI0
---.. 0
-.. . . . . . . . . . . . .
,
0
N7
0
.x ".
.'" .-..
o
S
~
.
NE ,
~ .... .
0
Cesi S.Martinofault~ll NEI ~ I
S17~
2km
I]
@
Fig. 3. (a) Geological cross-sections ($9, S12, S17) across the M. LeScalette and Cesi-S. Martino segments. Section $9 crosses M. LeScalette faults, section $12 crosses both M. LeScalette and Cesi-S. Martino faults, and section $17 shows the Cesi-S. Martino fault geometry. Location of the sections is given in Figure 2a. (b) Umbria-Marche stratigraphy with the position of the Marne a Fucoidi marker.
total throw achieved by the M. LeScalette faults reaches about 500 m. Section S12 crosses both the southeastern termination of fault N7, belonging to the M. LeScalette segment, and the northwestern termination of fault N l l , which belongs to the Cesi-S. Martino segment. Section S17 shows the geometry of the central part of fault N l l where the throw is about 550 m. The throw distribution along the M. LeScalette and Cesi-S. Martino faults was measured along the geological cross-sections, from the offset of a continuous stratigraphic marker (Marne a Fucoidi formationM F fm, Aptian-Albian) which is a marly formation (Fig. 3b) well recognisable within the well-known Umbria-Marche carbonatic succession (Cresta et al. 1989). Besides, the stratigraphic setting of the area allowed us to reconstruct the geometry of the MF marker by stratigraphic correlation where it does not crop out. The M. LeScalette segment has a maximum length of 9300 m and an asymmetric throw distribution with a maximum value of about 500 m about 3 km SE of the segment centre (Fig. 4). The Cesi-S. Martino segment has a maximum length of about 8200 m, and comprises a single major fault ( N l l ) with an asymmetric throw distribution. The maximum throw of about 550 m occurs at about 0.7 km to the
NW of the segment centre corresponding to section $18 (Fig. 4).
Comparing geological and morphological throws In tectonically active areas, faulting commonly offsets the topographic surface. Normal faults with nearly pure dip-slip kinematics, such as the CoFS faults, are visible in the landscape. The sum of the topographic offset and of the depth of basin infill (morphological throw) is often considered to be comparable with the long-term geological throw achieved by active normal faults (e.g. Roberts 1996; Coltorti & Pieruccini 2000; Burbank & Anderson 2001; Pizzi et al. 2002). This is especially valid in regions of low erosion rates (Doglioni et al. 1998). As the recent activity of these faults is recorded by variations in topographic elevations, it is expected that geological and morphological throws display similar values. This approach can be applied to the CoFS faults that are of Quaternary age and that differ from other normal faults of central Italy that reactivated pre-existing normal faults (Calamita et al. 1998; Pizzi & Scisciani 2000; Pizzi et al. 2002). Reactivation of a pre-existing normal fault was also suggested for the Gubbio
30
F. MIRABELLA 200
vw"
'
"
'
"
'
ET AL.
"
'
"
'
"
'
"
'
5 ~
S 1 ~ 1 5 LeScalettesegment
~ 1 5 o / ~ N l O f a u l t .
Cesi-SMartino
segment
511t) 1000 1500 20(x) 2500 3000 3500 4(K,Ud faultlengthtrn) 600 INW"
I
"
I
"
I
'
I
'
!
'
I
"
I
"
I
"
I
'SE
600 NW"
[.
"~
I
9
I
"
.
segment
r ~"F ~
St~
; 4001"-
... S~/
Itotal geological
~300~ S" ~
_
throw
3
~."
""
~ /~
~
-
c, ~ ~"F ~.
/
I
,~E400L
$17
. ~
,
~
sl
"~-- S~4"~N"N
~
~
I
9
S~ /
i
"
I
9
I
9
I
9
I
" SE
.
\
Cesi-S.Martino segment \
\ total geological
-
throw
.
~19
/
".~ .,~300I
io
I
s~7~s
I
~
.
$1r
-
N1~"~~23
9
.
~200
~l
-0
I1 - ~200
I
]
~
I
~
I
~ 1 ~
~ 3 ~
[ ~12
1000 2000 3000 4000 5000 6000 7000 8000 c ~ fault length (m)
24
~l
100013
0
I
I
I
[
~
am
l
[
l
[
I
I
I
I [
l
1000 2000 3000 4000 5000 6000 7000 8000 9000 fault length (m)
Fig. 4. Throw distribution along the studied fault segments and relative position of the fault segments in map view (top-left).
normal fault NW of Colfiorito (see location in Fig. la), from good quality seismic reflection profiles (Mirabella e t al. 2004). In such circumstances, erosion erased a considerable part of the previously achieved morphological throw, giving morphological throw values that are smaller than the geological throw. For the M. LeScalette and C e s i - S . Martino segments of the CoFS, we measured the morphological throw on 1 : 10 000 maps along the trace of the geological sections that we used to measure the geological throw values. In the Colfiorito area, the topographic throw is a good approximation to the morphological throw (topographic throw + basin thickness) as the intramontane basins are a few tens of metres deep. The deepest basin, Colfiorito, reaches a maximum depth of about 120 m only in its central part. Hence, the values of the morphological throw are given by the difference between the maximum elevation of the fault footwall and the minimum elevation of the hangingwall (Fig. 5). Morphological and geological throw curves (Fig. 6) are quite similar. The two are asymmetric with the highest values shifted toward the nearby fault segment. The similarity in both shape and values suggests the observed longterm geological throw accumulated during the
Quaternary. From the age of the intramontane basins (about 1.8 Ma), it is possible to calculate the average slip rate by dividing the faultparallel offset (displacement) by the duration of faulting (1.8 Ma). Assuming a mean fault dip of 65 ~ and nearly pure d i p - s l i p kinematics, a
SW
NE
-...
I
o ~
I GT.~] I I
Marne a Eucoidi hn
....
"
~
I
Fig. 5. Schematic representation of the geological and morphological throws measured on the cross-sections. In this area, the morphological throw can be considered as equivalent to the topographic offset (difference between the maximum and minimum elevations at faults' footwall and hangingwall respectively), as the basins infilment is of few metres, reaching 120 m only in the depocentre of the Colfiorito basin (Messina et aL 1999).
SEGMENTATION AND INTERACTION OF NORMAL FAULTS 600
NIC
I
'
I
"
I
"
I
'
I
'
I
"
I
"M.LeScalettesegment
5OO
~ 4OO
.
"
I
''
I
"SE
] 600
'VW'
~
$9
'
..... ~7~
...,.,\
-
300 -
~ 200 - ~ ' ~ ~ - ~
N
/
\
;,~ "~
~1l
t -I
I ' i 9I ' I ' I " I ' I
' SEI
mort~hOolwOgl~cSal
Cesi-S.Martinosegmen; t "~#~,f7 16- "~.~tot.algeologicalthrow 1
morpholoicalthrow...*PS8 ~
-
"
31
.." ~ ' ~
f
I- .." /
;sl t 2oo~7
......
1 "~is24 t
100 0
1000 2000 3000 4000 5000 6000 7000 8000 9000 1000000 fault length (m)
1000 2 0 0 0 3000 4000 5000 6000 fault length (m)
7000
8000'9000
Fig. 6. Distribution of geological and morphological throws of M. LeScalette and Cesi-S. Martino segments.
maximum throw of about 500 m gives an average slip rate of 0.3 mm/year, in agreement with average slip rates for this portion of the Northern Apennines (~
~
,
-~ u=
0
200 400 crack-sealnumber
100 200 300 crack-sealnumber
~ol[
~o 0 600
~,otl ~Olt
~SantaY.... ~ P l
El0
~ooI~ 0~
0
~
o oo
s
~
~4oo
.~ E
=
0
0
50 100 crack-sealnumber
~"
100 200 300 crack seal number
0 1 2 3 4 5 6 7 8 910 12 crack-sealthickness(micron) l ~_mooo,.~.~m, ....
~4
4o ~ol/
~
~-2o
2 0
0
oi_
~
~, 200
mean:4.1 micron~
o
50 100 crack-sealnumber
c
30.
-~ glo ~-
~
~
0 -' 0 100 200 300 crack-sealthickness(micron)
~20
~5
>9 loo I =
~ 10
~oo b- 15f j ~ ~
300/
2
~ ~Otlll
0 t ............... 0 500 1000 1500 crack-sealnumber
500 1000 1500 crack-sealnumber
. . . . .
0 100 200 300 crack-sealthickness(micron)
,o[i
~ 100
o 400/II J .~
o
200 400 600 crack-sealnumber
,oo
>~
mean:38 micron
g2Orll
~ ~oo
0
~ ~0 [ i i l l l ~ ~ 00 100 200 300 crack-sealthickness(micron) 40[ [ I I
~oo
4 x104
J
men44lmcron .........
_- 200
0
._s g
4~
10
~o 0
100 200 300 crack-sealnumber
0
0 5 10 crack-sealthickness(micron)
X 10 4
6 : f ~ X i
cI 7 6 0 0 r ~
In
E400 200
~=
0~
0
,
200 400 crack-sealnumber
0 ~
0
200 400 crack-sealnumber
~, 0 200 400 600 crack-sealthickness(micron)
?3
F. R E N A R D ET AL.
74 ..ICI 't-
.s
3
.o "t-
~ 2,5
-o 2.5
-~ 2.5
3
t:t
o
o
2
2
~- 1.5
-1.5
C
1.5
:6 6~ _o 0.5 0
100 200 300 S42abc crack seal No.
0
200 400 600 S42cdeg crack seal No.
1
1.4
..Q "C
'i=.
.-~ 1.2 "0
~
~o u'}
/'Y" J " ' -
1
f',
0,8
(-.n J
0.6
I
0.8 c.-
~0.5
0.4
0
500 1000 1500 S421hijm crack seal No.
3
~
2.5
~
2
g
1.5
0.6 0.2
= 0.4
r
m
r
'5 02 6~ o 0
I
0
._
50 100 SY3P crack seal No.
I
0 0.2 [
0
100 200 3OO SY32 crack seal No.
1 0
200 400 X45FXC crack seal No.
Fig. 6. Distribution of crack thicknesses. The data of the middle column in Figure 5 are sorted by decreasing thickness (rank-ordering technique). In a log-linear plot, the sorted data (black line) plot close to a straight line (red) over 80% of the signal (except for large and small cracks), indicating an exponential distribution. Residuals (blue line) have a mean value close to zero and no specific trend.
rank n. Represented in a log-linear diagram, the rank ordering plot is qualified by a straight line on more than 80% of each data set (Fig. 6). This plot represents the distribution function of the cracks in a direction parallel to the opening vector and this describes the cumulative distribution of the probability density function. The distribution is mostly linear in a log-linear diagram, indicating that the data can be described by an exponential function (Otnes & Enochson 1978) where the number n of cracks of thickness w obeys the relationship
equal to Wo. Table 1 shows that for all crackseal patterns, the mean crack thickness and the standard deviation are reasonably close. This measurement and the linear trends in Figure 6 indicate that the crack-seal data can be described by an exponential distribution, the deviation from an ideal distribution being small at small and large crack thicknesses. The large number of events (up to 1106 successive cracks for sample S421hijm) ensures that the exponential distribution is representative.
Fourier analysis: a b s e n c e of spatial correlation The characteristic scale Wo represents a physical length or arises through the dynamics of the fracturing process (Bonnet et al. 2001). Note that for the serpentine samples, the exponential distribution is not smooth, because the average crack aperture is close to the resolution of the optical measurement process. For an exponential distribution, the mean and the standard deviation of the data set should be
To study the spatial correlation of successive cracks, the Fourier transform of the binarized crack-seal patterns was calculated. In a l o g log plot, the Fourier spectrum shows a flat linear relationship (Fig. 7). This indicates that the crack-seal data behave like a random series, having no preferential correlation in the signal: the variations of crack thickness are not correlated spatially between each other.
STATISTICAL ANALYSES OF CRACK- SEAL PATTERNS 6
__u -t
gE 4 E O
75
6
I
......
- - initial data .. 9 randomly permuted dab
g
\1 ~.,.
4
initial data randomly permuted dab
E
E O
2
c-
- \
C
2
9
'-
0 0
O
u- 2 O
S42cdeg
S42abc
4
4
3
2
-
1
2
5
4
log of frequency (micron-1)
3
2
1
0
log of frequency (micron -1)
~ 6 ii[ initial data I .... randomly permuted dab
'5
~
E 4 E b
3
.
.
nitial data .
.
BI
E
~2 t-
"t-
"
0
~ o
-2
4
3
2
JSY3 1 3
1
log of frequency (micron-1) 5
\
I --:
2
1
0
log of frequency (micron-1) 6
initial data '
- -
daa
E E 3
initial data randomly permuted dab
E
2 t~
2
1 0 ii
ii . . . . . . . . . . . . . . . .
5Y32
or) O
--2
3
45FXC 2
1
0
log of frequency (micron -1)
2
5
4
3
2
1
0
log of frequency (micron -1)
Fig. 7. Fast Fourier Transform (FFT) analysis of crack-seal patterns. In log-log plot, the overall flat shape of the Fourier spectrum indicates that noise dominates the data and that no specific organization can be detected. The raw data (full lines) and the randomly permutated data (dotted lines) show a similar horizontal linear trend.
This interpretation was confirmed by performing Fourier transforms on randomly permuted sets of data. For each c r a c k - s e a l sequence, the positions of the cracks along the profile parallel to the vein opening direction were randomly permuted. This transformation gives series with the same m e a n and variance as the original ones and for which the initial order (spatial correlation), if any, is destroyed (dotted lines on Fig. 7). Performing Fourier transforms on these new sets
does not change the flatness of the Fourier spectrum. The question as to whether binarization of the original patterns taken with the camera could induce a bias on the high frequencies in the Fourier analysis needs to be addressed. To test this effect, Fourier transforms were also performed on the raw data, taken directly from the camera. Here again, the Fourier spectrum showed a flat linear relationship.
76
F. RENARD E T AL.
Another test, not shown here, is the autocorrelation function of the crack-seal sequence. For all samples, the autocorrelation function drops to zero immediately (as for randomly permutated data sets) thereby confirming that the crack-seal patterns are uncorrelated in space. Therefore, the successions of crack-seal thicknesses do not show any visible spatial organization. This is a memoryless process (the thickness of crack number n does not depend on the thickness of crack n - 1), which is also a property of exponential distributions.
Discussion To summarize the statistical analysis, cracks have an exponentially decaying thickness distribution (Fig. 5). The mean and standard deviation crack distribution values are close and the distribution is almost linear in a log-linear diagram (Fig. 6). Moreover the cracking events are not spatially correlated (Fig. 7). The crack-thickness distribution has all the properties of an exponential distribution. The exponential probability density function has been widely used to accurately describe the failure times in various systems (Fisz 1963; Otnes & Enochson 1978). It characterizes systems containing a large number of elements (transistors, elastic fibres, and so on) which break with a constant failure rate probability. As an analogy to these systems, the exponential distribution of cracks indicates that they are characterized by a single failure rate, which can be controlled, for example, by the external loading conditions. Geological systems such as mid-oceanic ridges (Cowie et al. 1993) also exhibit an exponential distribution. Based on numerical simulations on a spring network model, Spyropoulos et al. (2002) have concluded that exponential distributions of fractures in rocks are indicative of an increase in brittle strain in a regime dominated by the heterogeneity in the system. Moreover, Gupta & Scholz (2000) have observed a transition from a power-law to an exponential distribution of faults as brittle deformation increases in the Afar area. The three types of crack-seal veins show the same trend of statistical properties, regardless of the geological context, the type of rock and the vein kinematics (extensional or shear vein). The linear trend in the cumulative displacement plot (Fig. 5, left) is interpreted as being the fact that the far-field boundary loading conditions and the physico-chemical parameters during the crack-seal deformation did not vary notably; see also Davison (1995) for similar data. There
is neither acceleration nor slow-down in average vein growth rate. As a result, the variations observed in crack thickness should reflect disturbances around a steady-state fracturing process in a constant environment. This steady-state fracturing process is determined both by far-field stress conditions and by the local petrography and stress field in the wallrock. For example, the grain size in the wallrock could control the average crack thickness through nucleation processes or local variations in the wall-rock yield strength. Petrographic evidence also confirms that the P-T conditions did not change during vein formation. Intergrowths of quartz and tourmaline form the crack-seal pattern in the Abitibi quartz veins (Fig. 2). These two crystals have grown from hydrothermal conditions and mineral paragenesis and the fluid inclusions trapped in the successive layers indicate that the thermodynamic conditions (lithostatic pressure, temperature) have not changed much during the whole vein opening history (Firdaous 1995; Robert et al. 1995). Such observations are similar for the three types of veins. The crackseal sequence represents cycles of breakage events and periods of quiescence during which the filling minerals have grown from a fluid. During that time, stress and temperature were kept more or less constant. Consequently, crack-seal structures have recorded a sequence of events occurring in a rock domain with almost invariant geological conditions. In this respect, they have recorded small variations around a steady state. A crack-seal vein accumulates elastic strain due to tectonic loading until the stress on it reaches the yield strength. At this stage a new crack forms. The presence of a well-defined peak in the crackseal thickness distribution indicates that the yield stress varies around a mean value. The uncorrelated distribution of crack-seals and the exponential distribution of crack thicknesses can be interpreted by two effects. On one side, they can reflect spatial heterogeneities in the rock as a crack opens. On the other side, they can reflect time-effective stress and/or fluid pressure variations at the vein boundary. In the first assumption, the exponential distribution could be due to fluctuations in the crack length or in the local rock strength. The average crack length was measured in the field or directly on the samples (Table 1) and remains almost constant for all the cracks in each vein. In addition, each crack follows exactly the same parallel path as the previous one. Moreover there is no evidence that the crack thickness is related to the local mineralogy of the wall-rock.
STATISTICAL ANALYSES OF CRACK-SEAL PATTERNS For the Abitibi samples, the wall-rock contains centimetre-long quartz phenocrystals along which several tens of cracks have developed. Locally, these cracks also show an exponential distribution, even if there is no variation in the wall rock. Therefore there is no structural evidence in the sample that the heterogeneity content varies spatially and that there were local variations in the strength of the vein. In the second assumption, the crack thickness varies, whereas all the textural or geometric parameters of the veins remain unchanged. As the crack lengths are almost constant, the crack thickness variations reflect disturbances in the crack thickness/crack length ratio. Rock mechanics indicate that this ratio is directly related to the state of stress around a fracture and/or at a fracture tip. Given that, after the cracking event, the crack remains open, several authors have proposed that high fluid pressures open the cracks (Sibson et al. 1988; Boullier & Robert 1992). For a pressurized, elastically opened crack, crack width is proportional to the driving pressure (Pollard & Segall 1987). With this assumption, the overpressure or, necessary to open a crack of length c with an aperture w is
wE o- = - 4c
(2)
(Gu6guen & Palciauskas 1994). This corresponds to the elastic stress released when a new crack has opened. Crack surface areas are similar for all the individual cracks; only the crack thickness varies. In this case, the thickness can be used to calculate typical stress variations that allowed crack opening assuming that the crack length is known. What crack-seal structures record in fact are variations in elastic stress release during the vein formation. The values for c given in Table 1 represent a maximum length for the cracks as observed in the outcrops or directly on samples. Given these lengths and the typical elastic modulus of the wall-rocks, the variations in crack thickness indicate minimum effective driving pressure variations ranging from 10 to 50 bars. These variations are small compared to the main stresses. However, they are sufficient to explain the crack-seal patterns.
Conclusions Crack-seal veins, recording several hundred successive crack events, have been analysed. All data sets show similar statistical properties, irrespective of the various geological settings and vein kinematics: the exponential crack
77
distributions have many similarities with a random data set that has a well-defined characteristic length. In this way, the distributions are different from other fracture patterns, which obey power-laws (as in earthquakes) or other types of correlations. Crack-seal structures provide a fossil record of non-correlated crack sequences that have a well-defined average characteristic size: they have recorded noisy stress release variations in the crust. This project has been supported by the CNRS through an Action Thdmatique Innovante. We would like to thank C. Pequegnat and D. Tisserand for technical help. We thank S. Cox and P. Meakin for fruitful discussions and F. Robert for providing the Abitibi crack-seal samples. We acknowledge reviews by M. Jessell, X. Zhang and P. Cobbold.
References ATKINSON, B. K. 1984. Subcritical crack-propagation in geological materials. Journal of Geophysical Research, 89, 4077-4114. ATWATER, T. 1989. Plate tectonic history of the northeast Pacific and western North America. In: WINTERER, E. L., HUSSONG, D. M. & DECKER, R. W. (eds) The Eastern Paci[ic Ocean and Hawaii. Geology of North America, Geological Society of America, Boulder, CO, 21-71. BEACH, A. 1977. Vein arrays, hydraulic fractures and pressure solution structures in a deformed flysch sequence, S.W. England. Tectonophysics, 40, 201-225. BEUTNER, E. C. & DmGEL, F. A. 1985. Determination of fold kinematics fi'om syntectonic fibers in pressure shadows, Martinsburg slate, New Jersey. American Journal of Science, 285, 16-50. BONNET, E., BOUR, O., ODLING, N. E., DAVY, P., MAIN, l., COWIE, P. & BERKOWITZ, B. 2001. Scaling of fracture systems in geological media. Reviews of Geophysics, 39, 347-383. BONS, P. & JESSELL, M. 1997. Experimental simulation of the formation of fibrous veins by localised dissolution-precipitation creep. Mineralogical Magazine, 61, 53-63. BONS, P. 2000. The formation of veins and their microstructures. Journal of the Virtual Explorer, 2, 12. BOULHER, A. M. & ROBERT, F. 1992. Palaeoseismic events recorded in Archaean gold-quartz vein networks, Val d'Or, Abitibi, Quebec, Canada. Journal of Structural Geology, 14, 161 - 179. COWIE, P. C., SCHOLZ, C. H., EDWARDS, M. & MALINVERNO, A. 1993. Fault strain and seismic coupling on mid-oceanic ridges. Journal of Geophysical Research, 98, 17 911-17 920. Cox, S. F. & ETHERtDGE, M. A. 1983. Crack-seal fibre growth mechanisms and their significance in the development of oriented layer silicate microstructures. Tectonophysics, 92, 147-170.
78
F. RENARD ETAL.
Cox, S. F. 1987. Antitaxial crack-seal vein microstructures and their relationship to displacement paths. Journal of Structural Geology, 9, 779-787. Cox, S. F. 1991. Geometry and internal structures of mesothermal vein systems: implications for hydrodynamics and ore genesis during deformation, ln: Structural Geology in Mining and Exploration. University of Western Australia, Extended Abstracts, 25, 47-53. Cox, S. F. 1995. Faulting processes at high fluid pressures: an example of fault valve behavior from the Wattle Gully Fault, Victoria, Australia. Journal of Geophysical Research, 100, 12 841-12 859. DAVISON, I. 1995. Fault slip evolution determined from crack-seal veins in pull-apart and their implications for general slip models. Journal of Structural Geology, 7, 1025-1034. DILEK, Y., COULTON, A. & HURST, S. D. 1997. Serpentinization and hydrothermal veining in peridotites at site 920 in the Mark area. In: KARSON, J. A., CANNAT, M., MILLER, D. J. & ELTHON, D. (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 153, 35-59. ELLIS, M. 1986. The determination of progressive deformation histories from antitaxial syntectonic crystal fibers. Journal of Structural Geology, 8, 701-709. ELTER, P. 1973. Lineamenti tettonici ed evolutivi dell Appennino. Acad. Naz. Lincei, Roma, 183, 97-109. ERNST, W. G. 1971. Do mineral parageneses reflect unusually high pressure conditions of Franciscan metamorphism? American Journal of Science, 271, 81-108. FIRDAOUS, K. 1995. Etude des fluides dans une zone sismog~nique fossile: les gisements aurif~res m~sothermaux archdens de Val d'or, Abitibi, Quibec. University Thesis, Institut National Polytechnique de Lorraine. FISHER, D. M. & BRANTLEY, S. L. 1992. Models of quartz overgrowth and vein formation: deformation and episodic fluid flow in an ancient subduction zone. Journal of Geophysical Research, 97, 20 043- 20 061. FISHER, D. M., BRANTLEY, S. L., EVERETT, M. & DzvoYIK, J. 1995. Cyclic fluid flow through a regionally extensive fracture network within the Kodiak accretionary prism. Journal of Geophysical Research, 100, 12 881-12 894. FISZ, M. 1963. Probability Theory and Mathematical Statistics. Wiley Publications in Statistics, Wiley, New York. GAVIGLIO, P. 1986. Crack-seal mechanism in a limestone: a factor of deformation in strike-slip faulting. Tectonophysics, 131, 247-255. GUEGUEN, Y. & PALCIAUSKAS,V. 1994. Introduction to the Physics of Rocks. Princeton University Press, Princeton, NJ. GUPTA, A. & SCHOLZ, C.H. 2000. Brittle strain regime transition in the Afar depression: implications for fault growth and sea-floor spreading. Geology, 28, 1087-1090. HENDERSON, J. R., HENDERSON, M. N. & WRIGHT, T. O. 1990. Water-sill hypothesis for the origin of
certain veins in the Meguma Group, Nova Scotia, Canada. Geology, 18, 654-657. HILGERS, C. 2000. Vein growth in fractures: experimental, numerical and real rock studies. Unpublished PhD dissertation thesis, University of Aachen, Germany. LABAUME, P., BERTY, C. & LAURENT, P. 1991. Syndiagenetic evolution of shear structures in superficial nappes: an example from the Northern Apennines, NW Italy. Journal of Structural Geology, 13, 385-398. LABAUME, P. & RIO, D, 1994. Relationships between the subLigurian allochthon and the Tuscan foredeep turbidites in the Bobbio window (NW Apennines). Memorie della Societd Geologica Italiana, 48, 309-315. OTNES, R. K. & ENOCHSON, L. 1978. Applied Time Series Analysis. Wiley, New York. PAGE, B. M. 1972. Oceanic crust and mantle fragment in subduction complex near San Luis Obispo, California. Geological Society of America Bulletin, 83, 957-972. PAGE, B. M. 1981. The Southern Coast Range. In: ERNST, G. W. (ed) The Geotectonic Development of California. Prentice-Hall, Englewood Cliffs, New Jersey, 330-417. PASSCHIER, C. & TROUW, R. 1995. Microtectonics. Springer, Berlin. PETIT, J.-P., WIBBERLEY, C. A. J. & RUIZ, G. 1999. Crack-seal, slip: a new fault valve mechanism? Journal of Structural Geology, 21, 1199-1207. POLLARD, D. D. & SEGALL, P. 1987. Theoretical displacement and stresses near fractures in rock with applications to faults, joints, veins, dykes and solution surfaces. In: ATKINSON B. K. (ed) Fracture Mechanics of Rock. Academic Press, London, 277-350. RAMSAY, J. G. 1980. The crack-seal mechanism of rock deformation. Nature, 284, 135-139. RAMSAY,J. G. & HUBER, M. I. 1983. The Techniques of Modern Structural Geology. Volume 2: Folds and Fractures. Academic Press, London. ROBERT, F. 1990. Structural setting and control of gold-quartz veins of the Val d'Or area, southeastern Abitibi Subprovince. In: Ho, S. E., ROBERT, F. & GROVES, D. I. (eds) Gold and Base Metal Mineralization in the Abitibi Subprovince, Canada, with Emphasis on the Quebec Segment. University of Western Australia Publication, 24, 164-209. ROBERT, F. & BOULLIER, A. M. 1994. Mesothermal gold-quartz veins and earthquakes. In: HICKMAN, S. H., SIBSON, R. H. & BRUHN, R. L. (eds) The Mechanical Involvement of Fluids in Faulting. U.S. Geological Survey, Open-file report 94-228, 18-30. ROBERT, F., BOULLIER,A. M. & FIRDAOUS,K. 1995. Gold-quartz veins in metamorphic terranes and their bearing on the role of fluids in faulting. Journal of Geophysical Research, 100, 12 86112 879. ROBERT, F. & BROWN, A. C. 1986. Archean goldbearing quartz veins at the Sigma Mine, Abitibi greenstone belt, Quebec. Part I: Geologic relations
STATISTICAL ANALYSES OF CRACK-SEAL PATTERNS and formation of the vein systems. Economic Geology, 81, 578-592. SIBSON, R. H., RO~Er~T, F. & POULSEN, K. H. 1988. High angle reverse faults, fluid-pressure cycling and mesothermal gold quartz deposits. Geology, 16, 551-555. SPYROPOLOUS, C., SCHOLZ, C. H. & SHAW, B. E. 2002. Transition regimes for growing crack populations. Physical Review E, 65, 056105-1056105-10. STAMOUD1, C. 2002. Processus de serpentinisation des
piridotites de Hess-Deep et de la zone de MARK." Approches chimiques et miniralogiques. PhD thesis, University Paris VI, France. TABER, S. 1916. The growth of crystals under external pressure. American Journal of Science, XLI (4th series, No. 246), 532-556. TABER, S. 1918. The origin of veinlets in the Silurian and Devonian strata of central New York. Journal of Geology, 6, 56-73. TORIUMI, M. & HARA, E. 1995. Crack geometries and deformation by the crack-seal mechanism in the
79
Sambagawa metamorphic belt. Tectonophysics, 245, 249- 261. URAI, J. L., WILLIAMS, P. F. & van ROERMUND, H. L. M. 1991. Kinematics of crystal growth in syntectonic fibrous veins. Journal of Structural Geology, 13, 823-836. VAN DER PLUIJM,B. A. 1984. An unusual 'crack-seal' vein geometry. Journal of Structural Geology, 6, 593 -597. WILLIAMS, P. F. • URAI, J. L. 1989. Curved vein fibres: an alternative explanation. Tectonophysics, 158, 311-333. WILTSCHKO, D. V. & MO~SE, J. W. 2001. Crystallization pressure versus 'crack seal' as the mechanism for banded veins. Geology, 29, 79-82. Xu, G. 1997. Fluid inclusions in crack-seal veins at Dugaid River, Mount Isa Inlier: implications for paleostress states and deformation conditions during orogenesis. Journal of Structural Geology, 19, 1359-1368. ZIPF, G. K. 1949. Human Behavior and the Principle of Least-Effort. Addison-Wesley, Cambridge.
Single-contact pressure solution creep on calcite monocrystals SERGEY ZUBTSOV x'2, FRAN(~OIS R E N A R D 1'3, J E A N - P I E R R E G R A T I E R l, D A G K. D Y S T H E 3 & V L A D I M I R T R A S K I N E 2
1LGIT-CNRS-Observatoire, Universitd Joseph Fourier, BP 53, 38041 Grenoble cedex, France (e-mail: Francois.Renard@ lgit. obs. ujf-grenoble.fr) 2Colloid Chemistry Group, Chemistry Department, Moscow State University, Moscow 119899, Russia 3physics of Geological Processes, University of Oslo, Postboks 1048 Blindern, 0315 Oslo, Norway
Abstract: Pressure solution creep rates and interface structures have been measured by two methods on calcite single crystals. In the first kind of experiments, calcite monocrystals were indented at 40 ~ for six weeks using ceramic indenters under stresses in the 50-200 MPa range in a saturated solution of calcite and in a calcite-saturated aqueous solution of NH4C1. The deformation (depth of the hole below the indenter) is measured ex situ at the end of the experimenc In the second type of experiment, calcite monocrystals were indented by spherical glass indenters for 200 hours under stresses in the 0-100 MPa range at room temperature in a saturated aqueous solution of calcite. The displacement of the indenter was continuously recorded using a specially constructed differential dilatometer. The experiments conducted in a calcite-saturated aqueous solution of NH4CI show an enhanced indentation rate owing to the fairly high solubility of calcite in this solution. In contrast, the experiments conducted in a calcite-saturated aqueous solution show moderate indentation rate and the dry control experiments did not show any measurable deformation. The rate of calcite indentation is found to be inversely proportional to the indenter diameter, thus indicating that the process is diffusion-controlled. The microcracks in the dissolution region under the indenter dramatically enhance the rate of calcite indentation by a significant reduction of the distance of solute transport in the trapped fluid phase. This result indicates that care should be taken in extrapolating the kinetic data of pressure solution creep from one mineral to another.
Pressure solution creep (PSC), coupled with cataclastic and crystal plastic deformation, plays a significant part in sedimentary and fault rock deformation in the upper crust (Ramsay 1967). This is a mechanism of a slow aseismic compaction and porosity reduction whereby stressenhanced chemical potential at the contact sites of mineral grains results in local supersaturation of the adjacent fluid, diffusion of the solutes out of the high concentration areas, and precipitation of the material on the grain faces with smaller chemical potential (Weyl 1959; Paterson 1973; Rutter 1976). According to this definition, a distinction can be made between reactionkinetics-controlled PSC, where the rate of deformation is limited by the dissolution or precipitation of the material, and diffusion-controlled PSC, where this rate is limited by solute diffusion along the grain contacts (Raj 1982).
The main difficulty in quantifying which parameters control PSC rates is that this very slow mechanism of deformation is difficult to reproduce experimentally. Moreover, this process largely depends on the shape of the solid that evolves with time because of deformation. This is the main reason why indenter experiments have been developed, where an inert indenter in contact with a crystal and its solution is shown to pressure dissolve the crystal (Gratier 1993; Dysthe et al. 2003). As the shape of the indenter does not change with time, the deformation is easier to quantify. Moreover, the deformation laws for this simple geometry are well known and allow separating the different parameters controlling the rate of deformation. By this means one expects to deliver creep laws for relevant geological conditions and to quantify both finite strain and rates of PSC deformation.
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and
Tectonics:from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 81-95. 0305-8719/05/$15.00 (~ The Geological Society of London 2005.
S. ZUBTSOV ET AL.
82
The displacement rate of a piston (~/) indenting a mineral by PSC is deduced from theoretical laws (Raj 1982; Rutter 1983) for the two types of PSC. For the diffusion-controlled case, the relationship is Ocw
~oc d 2
(1)
where c is the mineral solubility at stressed state, w the thickness of the water film under the indenter, D the diffusivity of rate limiting ion, and d the indenter diameter. Here, C:
CoeAIx/RT
(2)
where Co is the mineral solubility at zero stress, R the gas constant, T the temperature, and A/z the difference in chemical potential between the dissolution and precipitation zones (proportional to the applied normal stress O-n). At low A/~ (if o'n < 3 0 MPa; Rutter 1976) an approximation may be used to obtain the relation derived by Gratier (1993):
Dco w A l~ RTd 2
(3)
For the kinetic-controlled case, the relationship is
yoc k(C - C~ \ co /
(4)
or expressed more simply:
;y oc kc
(5)
Here ( c - Co)/Co is the saturation state of the solution and k the reaction rate constant of the interfacial reaction (dissolution or precipitation). For precise calculations it should be borne in mind that this constant is temperaturedependent. Using relation (2) at crn < 30MPa, the relation derived by Gratier (1993) for mineral indentation by kinetic-controlled PSC is obtained, namely:
kcoAl~ RT
(6)
However, relation (4) is based on the approximation that the rate of interfacial reaction is proportional to the saturation state of the adjacent solution. This is not always true. In the general case, this rate (and thus the rate of mineral indentation by kinetic-controlled
PSC) is in power relation with the saturation state (Morse & Arvidson 2002):
:y oc k(C - C------~~n \ c0 /
(7)
At low applied stress, a relationship similar to that of de Meer and Spiers (1999) can be derived from equation (7):
kco( A tz) n RT
~/o~ -
(8)
Here n is the empirical reaction order, which is different for different minerals. Moreover, it is different for different solution compositions and usually different for dissolution and precipitation processes (Morse & Arvidson 2002). For calcite in pure water, n was found to be equal to 1 for dissolution (Plummer et al. 1978) and 2 for precipitation (Reddy & Wang 1980). Generally, all of the preceding means that the relationship between the rate of mineral indentation by PSC and the value of the difference in chemical potential between the dissolution and precipitation zones (i.e. the value of applied normal stress On) is sophisticated and depends on many factors. Care should be taken when using the chemical potential to determine whether kinetics or diffusion-controlled PSC takes place in an experiment. On the other hand, equations (1) and (5) show that, if the process of crystal indentation is diffusion-controlled, the rate of indenter displacement is inversely proportional to the square of the indenter diameter, and in the case of kinetic control the indenter diameter has no influence on this rate. The relationship between indenter diameter and indentation rate seems to be a more reliable parameter on which to base conclusions concerning the controlling mechanism of PSC. Two types of laboratory experiment have been conducted previously to investigate PSC. In the single contact experiments, the crystal surface is indented by a piston or another crystal (Tada & Siever 1986; Gratier 1993; Dysthe et al. 2002). This type of experiment is conducted in order to observe the deformation process precisely controlling the applied normal stress, the contact surface area and the displacement of the indenter. These experiments represent a useful tool for validating existing theoretical pressure solution models (Dysthe et al. 2003). Unfortunately, for low-solubility minerals, such as calcite or quartz, the rate of single contact deformation is too slow and the laboratory experiment must last several months or the displacement recording system must be extremely sensitive. This is why a second experimental
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS approach is used for these minerals. It consists of powder compaction, that is, multiplying the number of grain contacts to increase the bulk deformation rate (de Meer & Spiers 1997; den Brok et al. 1999; Zang et al. 2002). However, the interpretation of the results obtained is ambiguous as other deformation processes (grain sliding, subcritical crack growth) can occur in addition to PSC (Zang et al. 2002). Contrary to the significant number of theoretical and experimental studies of cataclastic and plastic calcite deformation, the PSC of calcite has not previously been studied in great detail in spite of its importance in natural rock deformation processes. The main reason for this situation is the low solubility of calcite, which makes it difficult to investigate its PSC in laboratory conditions. However, a limited number of pressure solution compaction experiments using calcite powder have been performed. Baker et al. (1980) studied compaction of deep-sea carbonate Iceland spar and reagent-grade calcite powder at 3 5 - 1 0 0 M P a and 2 2 - 1 8 0 ~ The duration of experiments varied from 21 to 240 hours. After 240 hours of compaction, a porosity reduction of 9.9% was obtained. Unfortunately, there is no clear evidence in these experiments that cataclastic and crystal plastic deformation do not make a significant contribution to the porosity reduction value obtained. Recently, Zang et al. (2002) compacted finegrained calcite powders at ambient temperature. The compaction experiments were performed in drained conditions with a saturated calcite solution, applying a uniaxial effective stress of 1 - 4 MPa. All samples were initially precompacted at 8 MPa to minimize the contribution of grain sliding and reorientation to subsequent wet deformation. The absence of brittle deformation during the experiment was proven by the absence of recorded acoustic emissions. However, a mechanism of subcritical crack growth was considered to be a relevant deformation process. Dry experiments and experiments with inert fluid did not show any significant deformation. Deformation of the wet samples reached 1.1% maximum strain after 20 days of compaction. The compaction rate decreased with increasing strain, increasing grain size and decreasing applied stress. On the basis of these data, it was concluded that reaction kinetic-controlled pressure solution creep and possibly subcritical crack growth were the dominant deformation mechanisms. To our knowledge, no single contact PSC experiments have been performed to date because of the very slow rate of calcite single crystal deformation. There are at least three different approaches
83
described below to resolve this problem: modifying the fluid chemistry, inducing microcracks, and increasing the sensitivity of the deformation measurement device. Effect o f f l u i d chemistry
The nature of the fluid is an important parameter for the rate of PSC. Above all, it is obvious that the rate of deformation depends on the solvent capacity of the fluid; that is, it is proportional to the solubility of the mineral in the fluid. For example, Pharr & Ashby (1983) studied the rate of pressure solution deformation of precompressed potassium chloride and sucrose powders in methanol/water solutions. They found that the extent of deformation enhancement is a linear function of the solubility of the solid in the pore fluid. Gratier & Guiguet (1986) used a solution of NaOH to increase the deformation rate of quartz, starting from the idea that alkalinity enhances its solubility as well as its dissolution and growth kinetics. Moreover, the chemistry of the fluid can influence the rate of pressure solution deformation by phenomena such as adsorption of fluid components and complex formation. Skvortsova et al. (1994) investigated the effect of adding small quantities of dimethylaniline (DMA) on the rate of deformation of halite by PSC. They found that DMA forms complexes with salt and adsorbs on the salt crystal surfaces, resulting in a change from a diffusion-controlled to a reaction-controlled mechanism. This transition is accompanied by a sharp decrease in deformation rate. In addition, Zang et al. (2002) found that calcite powder PSC slows down significantly if a small quantity of MgC12.6H20 is added to the pore fluid. The reason for this effect is the adsorption of magnesium ions on the calcite crystal surfaces, which inhibits further calcite crystallization. The rate of calcite deformation by PSC can also be enhanced by changing the fluid chemistry. As calcite is a weak base, its solubility is increased in acid solutions. In particular, in weak acid solutions, a dynamic chemical equilibrium takes place so that calcite cannot only be dissolved in the solutions, but also precipitates if the solution is supersaturated. This is not the case for a total dissolution reaction, as for example when calcite dissolves in HC1. A wellknown natural example of this dynamic equilibrium is the dissolution of calcite in a solution saturated in carbon dioxide: CaCO3 q- C O 2 "q-H20 = Ca(HCO3) 2
(9)
The rate of mineral deformation by PSC is proportional to its solubility; consequently the rate
84
S. ZUBTSOV ET AL.
of calcite deformation by PSC should be enhanced in a weak acid solution saturated by calcite. Effect of microcracks Gratier et al. (1999) have shown in experiments of halite indentation by diffusion-controlled PSC that if the applied stress is high enough to form microcracks on the mineral surface under the indenter, the rate of deformation enhances dramatically. The reason for this effect is that the kinetics of diffusion along a trapped fluid phase under the indenter could be one order of magnitude less than in bulk fluid (Dysthe et al. 2002; Alcantar et al. 2003). If there are no microcracks under the indenter, the indentation rate is inversely proportional to the square of the indenter diameter. However, in the presence of microcracks, faster diffusion out of the contact may occur in the free fluid that fills the microcracks. Consequently, in this case the rate of mineral indentation is not limited by the diameter of the indenter, but by the distance between two adjacent microcracks. It should be noted that calcite is a brittle mineral and the formation of microcracks on its surface under the indenter seems plausible. Increase in recording sensitivity Another possibility for carrying out a singlecontact PSC experiment with calcite is to increase the sensitivity of the displacement recording system. Dysthe et al. (2002, 2003) developed a capacitance dilatometer especially designed for single-contact pressure solution experiments. The sensitivity of this device is about 10 - 9 m. Halite monocrystals were tested using the dilatometer and showed a creep rate of about 2 x 10 - 6 to 10 - 8 m/day under an applied stress of 4 MPa (Dysthe et al. 2003). In this contribution, two types of singlecontact PSC experiments with calcite monocrystals are described. The three abovementioned ways of activating PSC on calcite crystals were successfully used in this study. The interpretation of the results obtained provides new constraints for understanding the mechanism of PSC on calcite deformation.
Experimental methods P S C indenter experiments" with ex situ deformation measurement With the simple indenter technique described in detail by Gratier (1993), calcite monocrystals
were indented at 40 ~ using ceramic indenters under stresses in the 50-200 MPa range for six weeks in a saturated solution of calcite and in a calcite-saturated aqueous solution of NH4C1 (Fig. la). Natural calcite crystals (from Sweetwater Mine, Reynolds Country, Missouri, USA, delivered by Ward's Natural Science Establishment Inc., Rochester, NY, USA) were cleaved and then polished (1200 to 4000 mesh grinding paper and DP-mol cloth with 1 micron diamond suspension, all from Struers A/S, Rodovre, Denmark). The surfaces were cleaned with water. The aqueous solutions were checked to ensure that they wetted the crystals when the experiments were started. For some experiments without water, the calcite crystals were dried with nitrogen. For each experiment, a different crystal was used and the indenter was always applied on the same crystallographic face. The saturated solution of calcite was prepared by adding several grams of CaCO3 powder to 100 ml of distilled water boiled for 15 minutes before solution preparation in order to eliminate any traces of CO2, and heated to 60 ~ shaking once per day for several days. The aqueous calcite-saturated solution of NH4C1 was prepared by adding 10 g of CaCO3 to 100 ml of a 5% solution of NH4C1 in distilled water and heating to 60 ~ shaking once per day for several days. Before starting the experiments, the solutions were kept at 40 ~ (temperature of the experiments) for two days. Each deformation cell consists of a Plexiglas tube with a diameter of several centimetres, and a Plexiglas bottom. The ceramic indenters, 300 txm wide and of various lengths (see Table 1), are mounted under a free-moving Teflon | piston and put in contact with calcite monocrystals immersed in a saturated solution. Dead weights placed on the piston maintain a constant stress on the samples. The value of the dead weight is calculated for each indenter in order to impose the desired stress (Table 1). Liquid paraffin is added on the top of each cell to prevent evaporation of the fluid. The cells are placed in a furnace at constant temperature. After six weeks, the depths of the resulting holes are measured. Experimental conditions (stress, length of indenter, nature of fluid) for different deformation cells, together with the depth of the holes obtained, are presented in Table 1. The indenter lengths are different for each indenter (from 300 to 740 txm), but the diffusion distance of solutes under the indenter is always the same because of the same indenter widths (300 txm).
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS
1
2
aluminium block \
85
3
:lead a~au
Peltier elements
brass block
9
+/ iOmm
, I
"
isolation
J
_ . . . . . . s . . . . . . . . . . . . . . . . . . . . . . . . . . . I t .. . . . . . . . . . . . . . . . . . . . s !
,
"
oil ,
dry sample
'
upper ~ ~ ~ i n d e n ' e r i / / electr ...................... ............[ t / / ~ c e n t e r e l e c t r o d e i , . i .~i}
'\
soluiion
" wet sample " t ' - ' - ' ~ sapphire window
~ / f - - ~ \ \ ~\
--(
O
~-~
~'/ glass spheres glued on the indenter
Fig. 1. (a) PSC indenter experiments with ex situ deformation measurement. First stage: equilibrium between calcite crystal and its saturated solution. Second stage: the indenter is placed on the mineral surface and vertical stress is applied through a dead weight. Third stage: slow creep of the indenter downwards into the crystal. The deformation is analysed ex situ by measuring the depth of the imprint. (b) High-resolution PSC experiments with continuous displacement recording through a double capacitance device (Dysthe et al. 2002, 2003). The technique allows displacement to be measured accurately with a long-term resolution of 2 nm (see text).
86
S. ZUBTSOV ET AL.
Table 1. Experimental conditions and results for PSC indenter experiments with ex situ deformation measurement. All the experiments last for six weeks at constant temperature (40 ~ All the indenters have a constant width (300 lzm) and varying lengths
Table 2. Experimental conditions for high-resolution PSC experiments with continuous deformation recording
No.
1
1 3 4 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 26 27 28 29 30 31 32 33 34 35 37 38 40
Length Dead Stress of the weight (MPa) indenter (kg) (~m) 300 380 360 310 320 500 340 300 400 400 300 320 300 300 300 360 740 380 300 340 340 320 380 300 400 300 300 300 300 360 340 390 360 340 300
1.20 1.52 1.44 1.24 1.28 2.00 0.77 0.68 0.90 0.90 0.68 0.72 0.68 0.68 0.30 0.36 0.19 0.10 1.20 1.36 1.36 1.28 1.52 1.20 1.60 0.68 0.68 0.68 0.68 0.81 0.77 0.88 0.36 0.34 0.08
200 200 200 200 200 200 150 150 150 150 150 150 150 150 100 100 50 50 200 200 200 200 200 200 200 150 150 150 150 150 150 150 100 100 50
Solution Depth chemistry of hole (txm)
No.
2 3
Calcite Calcite Calcite Calcite Calcite Dry Dry Calcite Calcite Calcite Calcite Calcite Calcite Calcite Calcite Calcite Calcite Calcite NH4C1 NH4C1 NH4C1 NH4C1 NH4CI NH4CI Dry Dry NH4C1 NH4C1 NH4CI NH4C1 NH4C1 NH4C1 NH4C1 NH4C1 NH4C1
0 0 0 5 5 0 0 19 0 5 5 18 5 0 0 0 5 5 52 65 29 48 108 46 0 0 15 4O 12 23 5 42 25 16 5
High-resolution P S C experiments with continuous deformation recording Experiments were conducted using the differential dilatometer described in Dysthe et al. (2002, 2003), which allows displacement to be measured accurately with a long-term resolution of 2 nm (Fig. l b). The device consists of two nominally identical capacitances, indenters and samples, one wet and one dry. This symmetrical design serves to eliminate all effects except those caused by differences on the two sides (presence/absence of water). The temperature is controlled by thermostatically regulated water
Fluid chemistry
Dead weight (g)
Saturated aqueous solution of calcite Saturated aqueous solution of calcite Dry (dried with nitrogen, immersed in hexadecane)
0; 80; 167 167 167
flowing through brass tubes spiraling down the external part of the instrument. The electronic circuit is a specialized lock-in amplifier built in-house and based on that of Jones (1988). About 2 0 - 3 0 glass spheres with diameters ranging from 250 to 300 txm were glued along the edge of a cylindrical glass support. This kind of indenter geometry provides good mechanical stability during the experiment. As the experiment goes on, the spheres penetrate into the 1.5 mm diameter calcite crystal. The contact surface area between the indenter and the crystal gradually increases, resulting in a progressive decrease in stress, The samples and solutions were prepared as described for the long-term calcite monocrystal indentation experiments. Table 2 shows the experimental conditions for these experiments. All experiments were run at stabilized room temperature. The temperature was controlled with circulating water from a thermally controlled bath, with fluctuations of about _+0.1 ~ that are larger than that described in Dysthe et al. (2003) for pressure solution on halite. Because the solubility of calcite is lower than halite and not very dependent on temperature (0.07% change with 0.1 ~ temperature change) the calcite experiments are not as sensitive to temperature fluctuations as the halite experiments. From time to time the electrical noise in the environment of the instrument causes short fluctuations in the measured indentation. These fluctuations make quantitative interpretation difficult for some parts of the experimental curves.
Results P S C indenter experiments with ex situ deformation measurement Figure 2 shows a typical hole obtained after calcite monocrystal indentation. The depth of holes was measured after the experiments using
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS
87
120
100
S.
80
60
,.'"
40
50
Fig. 2. Typical hole in a calcite monocrystal obtained after six weeks of indentation. The shape of the hole, close to a parallelepiped, represents that of the indenter. The depth of each hole is measured under microscope at the end of the experiment. a microscope with a resolution of three microns. The results are presented in Table 1. This table shows that none of the dry experiments (experiments 8, 9, 28, and 29) show any trace of deformation even under a maximum stress of 200 MPa. The experiments conducted with saturated calcite solution also usually do not show any deformation or show imprints of more than 5 ~m depth after six weeks (experiments 1, 3, 4, 6, 7, 11-13, 15-20). Depths greater than 5 ~m have been observed in two experiments only (10 and 14), which can be interpreted as brittle deformation. Therefore, it may be concluded that, under these conditions, there is no significant calcite deformation if saturated calcite solution is used as a fluid. On the other hand, the experiments conducted in a calcite-saturated aqueous solution of NHaC1 show significant deformation (Table 1, experiments 21-24, 26, 27, 30-35, 37, 38, 40). Moreover, these data indicate a correlation between the depth of the hole and the applied stress (Fig. 3).
High-resolution PSC experiments with continuous displacement recording Experimental data. Figure 4 shows the results of the first PSC experiment with the double capacitance dilatometer. During this experiment the load was varied as shown in Figure 4a. In this experiment the vertical displacement of the indenter is recorded continuously. This displacement corresponds to the indentation of some glass spheres in the calcite crystal. The
100 150 Stress, MPa
200
250
Fig. 3. Relationships between depth of hole and applied stress in the long-term indentation experiments conducted in a calcite-saturated aqueous solution of NHaC1. The straight line represents a guide for the eye.
temperature during the experiment is shown in Figure 4b. Figure 4a and b show that differential measurement makes the double capacitance insensitive to temperature fluctuations of the order of 0.1 ~ The disturbances in the indentation curve during the first five hours of the experiment arise from the failure of the temperature control system (see Fig. 4b). Figure 4a shows that there are large short-term fluctuations in the measured indentation. These are caused by electrical noise in the environment of the instrument. However, it is possible to separate the rapid jumps due to noise from the slowly varying capacitance due to indentation by PSC. Table 3 shows the indentation rates obtained by linear regression to the noise-free periods of the indentation curve shown in Figure 4a. The rate varies from 6 to 47 nm/h. It is believed that these variations represent the changes in stress under the indenter and the dissolution of minor asperities of calcite. However, the most important result of this experiment is the possibility of measuring the rate of calcite deformation with a resolution never reached before and to obtain relevant indentation rates. As illustrated in Figure 4c, an increase in dead weight by a factor of 2.1 (from 80 to 167 g) has an immediate effect on the deformation rate, which increases by a factor of 2.7 (from 11 nm/h in the interval 3 8 - 4 6 h to 30 nm/h in the interval 4 6 - 5 4 h ) . Figure 4d also shows that if the dead weight is removed, the indentation process stops completely. The electrical disturbances make it difficult to view this effect clearly throughout the whole stress-free period.
88
S. ZUBTSOV ETAL. 3 E
~2
o l/j
1
0g
167g
67 g
~ o~
._
o
~o
~o
20
40
6o
80
60
80
100 1'20 time, h
1~40
1'60
1'80
200
140
160
180
1~40
160
180~00
45 40 ~5 L. -~ 35
L)
II ~ 26,6
~3o III ~
100
l
E N 25 J L _ J
120
time, h
2 0 ~ 0
60
80
1130 1120 time, h
E =-1.7 E el.E E d9
f
J
~1.~
Q_
~9 1.4
"O
167 g
1.!
8
,~0
42
,;4
~e
,~8
time, h i
go r
g2
54
1'15
116
i
~2.21~
2.2~ E
2.2z
~Q'212z ,_ -(3
2.2
/ 09
1'10
1'11
i
112
1'13
i
114
timo., h
Fig. 4. Experimental results of the first high-resolution pressure solution creep experiment. (a) Indentation curve: the dead weight was changed several times in the course of the experiment as shown on the figure. (b) Temperature during the experiment. (c, d) Zooms on the creep curve.
Nevertheless, the moderately long interval without electrical disturbances and without any movement of the indenter can be distinguished in Figure 4d (Table 3, interval 109.5-113.3 h). The indentation process starts again with the same rate as soon as the dead weight is replaced in its original position (Table 3, intervals 7 9 - 8 9 and 122-128 h). The results of the second experiment are shown in Figure 5a. The only difference from
the first experiment is a new indenter with a different number of spheres touching the crystal surface. Figure 5b shows the temperature of the experiment. The rates of indentation in the noise-free parts of the signal are 48 and 56 n m / h at the beginning and end of the experiment, respectively. Figure 6a shows the results of calcite dry indentation. The indentation curve only shows some electrically induced fluctuations between
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS Table 3. Rates of deformation at different time intervals of the first high-resolution PSC experiment (see Table 2) Time (h)
Indentation rate (nm/h)
Dead weight (g)
23.3 46.6 15.0 27.6 18.3 14.6 0.0 15.4 6.5
80 80 80 167 167 167 0 167 167
9-15 19-24 30-35 48-54 69-79 79-89 109.5-113.3 122-128 169-182
envelope and thus the relative height of the imaged surface at each pixel is determined with a resolution of 3 nm. The horizontal resolution depends on the lens used and with the highest magnification it is at the diffraction limit of about 0.5 pxm. The calcite crystal used in the first experiment was examined using this method. Figure 7a shows an indented part of a crystal surface. This image shows all the spherical imprints of the indenter and the scratches produced by the horizontal movement of the indenter. The depth of the holes is about 0.65 Ixm and that of the scratches is less than 0.30 ~m. These observations show that the change in dead weight during the experiment causes minor displacement of the indenter accompanied by surface scratching. When the spheres are placed under stress again, they restart crystal indentation until a new change in dead weight. This is the reason for the moderate depth of the holes measured by WLI, comparable to the values of deformation obtained in the experiment. Figure 7b presents a typical spherical dissolution imprint and its two-dimensional profile. The depth of the hole is about 1 o,m and its diameter is approximately 50 txm. The hole is surrounded by rings, which are about 200 nm higher than the crystal surface. These rings represent the reprecipitated calcite, Observations also indicate that the holes are produced by the
0 and 200nm, which are 10-15 times less than the deformation values obtained in the wet experiments for the same period of time.
Profiles of indented regions. The sample surfaces with indented regions are imaged by white light interferometer (WLI) microscopy (Wyko 2000 Surface Profiler from Veeco). The profiler is a microscope with a reference arm creating interference fringes with maximum intensity at equal optical path lengths of the imaging beam and reference beam. By vertical movement of the sample and simultaneous image capturing, the interference intensity
12 E
89
F
10
E
8
u9
6
~ 4
._
2 0 0
L
I
~
I
I
20
40
6O
8O
100
26.9
L
140
i
i
i
T
i
26.6
~
26.75
160
--
%
26.85
~
&
i
120
time, h
a
(J .
_
26,7
210 0
~ 40
J_ 60
, 80
i 1 O0
~ 120
140
160
time, h
Fig. 5. Experimental results of the second high-resolution pressure solution experiment. (a) Indentation curve; (b) temperature during the experiment.
S. ZUBTSOV ET AL.
90 0.2 E =.0.15 0.1 ~.. 0.05 ._
0 0
1'0
:~0
I
30
a
I
time, h
40
I
~0
60
26.7 ,o ~- 26.6s ~_ 26,61~
"~
E 26.64
.,.-,
26'620 b
1'0
20
I
I
30
40
50
dO
time, h
Fig. 6. Experimental results of the third (dry) high-resolution pressure solution experiment. (a) Indentation curve; (b) temperature during the experiment.
dissolution of calcite and not by subcritical crack growth and consequent brittle deformation. There is also evidence that some cracks formed under the indenter along the calcite cleavage planes. These cracks are produced by the action of high stress on the brittle mineral.
Stress estimate. For further calculations, the stress applied on the crystal surface in our experiments should be estimated. Given that the dead weight is 167 g, the diameters of the spherical imprints on the crystal surface are about 50 txm, the number of spheres touching the crystal surface is around 10, the stress could be assessed at 100 MPa.
Discussion Dominant deformation mechanism PSC indenter experiments with ex situ deformation measurement. The absence of any imprints on the crystal surface in the dry experiments strongly suggests that there is no crystal plastic deformation of the crystals and that the dominant mechanism is water-assisted deformation. The presence of dissolution imprints suggests that PSC is the main deformation mechanism.
High-resolution PSC experiments with continuous deformation recording. The deformation process stops in the absence of dead weight, that is, when no more stress is applied to the crystal surface. This fact indicates that, in the experiments conducted, deformation is effectively being measured and not electrical fluctuations. The spherical imprints of the indenters on the crystal surface (Fig. 7) indicate that the deformation measured is the deformation of the crystal surface itself and not of parts of the experimental device. The absence of any significant deformation in the dry experiments strongly indicates that the dominant mechanism of deformation is pressure solution. The presence of nearby grown calcite around the indentation imprint (Fig. 7b) also confirms this conclusion. Rate limiting process The rate of calcite monocrystal indentation in the high-resolution PSC experiments is about 1 jxm/ day for a stress of 100 MPa (see Table 2 and the results of the second experiment). On the other hand, the experiments with ex situ deformation measurements did not show any significant deformation after 42 days of indentation, even under a maximum stress of 200 MPa. The difference between these two types of experiment is the length of the diffusion path under the indenter which is six times shorter in the high-resolution
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS
91
Fig. 7. White light interferometry image of the indented surface of a calcite monocrystal from the first high-resolution pressure solution creep experiment. (a) Spherical imprints of the indenter and some scratches produced by the horizontal movement of the indenter on the whole calcite crystal. (b) Zoom on a typical imprint of a sphere on the calcite surface and its cross-sections, which can be used to determine the depth of the imprint and also shows the 200 nm height rims of calcite precipitation near the edge of the indentation (arrows).
PSC experiments than in the experiments with e x s i t u deformation measurement (the diameter of the spherical imprint on the crystal of the high-resolution experiment is 50 ~zm compared to the 300 Ixm width of the indenters in experiments with e x s i t u deformation measurement). Given that the rate of crystal indentation is independent of indenter diameter in the case of kinetic-controlled PSC (equation 4) and inversely proportional to the square of the indenter diameter in the case of diffusion-controlled PSC (equation 3), it could be suggested that diffusion is the limiting step in our experiments. To test this assumption, the rate of crystal indentation in the experiments with e x s i t u deformation measurement should be calculated for the diffusion-controlled case on the basis of the rate
in the high-resolution experiments. Starting from the difference in the length of diffusion paths, it is easy to see that this rate should be about 0.03 Ixm/day multiplied by a factor that represents the difference in the stress in these two experiments. This means that the final deformation after the end of the experiment (42 days) should be no more than several microns. This value is in good agreement with the experimental result - the absence of any measurable deformation after the end of the experiment. Thus, it may be concluded that calcite indentation by PSC under our experimental conditions is limited by the diffusion of solutes under the indenter. The diffusion control of calcite PSC in our experiments is a crucial difference between our
S. ZUBTSOV ET AL.
92
results and those of Zang et al. (2002). The plausible explanation of this difference is the indenter technique used here compared to the powder compaction method. Because of the very slow rate of calcite deformation by PSC, the maximum value of powder compaction obtained was only 1.2%. This means that the contact interface areas between two adjacent grains remain very small and irregular throughout the experiment. Therefore, the diffusion distances in this experiment are always extremely short. This may be the reason for kinetic control operating in the calcite powder compaction experiments. On the other hand, in the indentation experiments, significantly larger diameters of indenter impose a longer diffusion path, and the process is thus diffusion controlled.
Effect of the mineral solubility The experiments with ex situ deformation measurement with a saturated aqueous solution of calcite do not show any significant deformation. On the other hand, in the experiments conducted in a calcite-saturated aqueous solution of NH4C1, deformation of several tens of microns can be observed (Table 1). The reason for this change probably is the effect of ammonium chloride as a weak acid on the solubility of calcite. Solutions of weak acids should increase the rate of calcite deformation by PSC. For example, in an aqueous solution of NH4C1 saturated by calcite there is a dynamic equilibrium: 2CACO3 + 2NH4C1 + 2H20 = Ca(HCO3)2 + CaC12 + 2NHaOH
(10)
The chemical potential of the CaCO3 surface under the indenter is higher than that of the free surface. Therefore, in this region, the equilibrium is displaced to the right-hand side of equation (10) and dissolution of calcite takes place. This additional quantity of dissolved calcite diffuses to the free surface, where the concentration of dissolved calcite is lower, and precipitates there. The solubility of calcite in a 5% solution of NH4C1 is about 0.5 g/1 (Lafaucheux 1974), that is, 30 times higher than in pure water. This means that in this solution the final depth of the holes obtained after the end of the experiment should be 30 times higher than in the case of pure calcite solution. Based on the depth value of several microns in the case of pure calcite solution (see calculations in previous section), the depth of holes in the presence of NH4C1 may be estimated to be several tens of microns. This value is in agreement with the experimental results.
Effect of stress PSC indenter experiments with ex situ deformation measurement. Unfortunately, the scatter of the data (Fig. 3) makes it impossible to obtain a good approximation of this correlation by any simple function. On this figure, a straight line is represented as a guide for the eyes, but a power-law or an exponential function could also fit the data. It may be only noted that enhancement of the applied stress increases the rate of mineral indentation by PSC. High-resolution PSC experiments with continuous deformation recording. Figure 7a shows that the change in dead weight during the experiment causes some minor horizontal displacement of the indenter. When the spheres are under stress again, they restart crystal indentation until a new change in dead weight. During the first minutes of indentation, the contact area between the spheres and the surface is extremely small because of the small irregularities that are present even on a very well polished surface; therefore the spheres apply an excessively high stress on new areas of the crystal surface. As a consequence, a dramatic increase in deformation rate should be detected during the first minutes after the change in dead weight. This effect is clearly seen on the experimental curve when the dead weight changes from 0 to 167 g at t = 144 h (Fig. 4d). On the other hand, this is not the case when the dead weight changes from 80 to 167 g at t = 46 h (Fig. 4c). Maybe, in this second example, the indenter did not move horizontally during the change in dead weight and each sphere stayed in its initial hole. An increase in dead weight by a factor of 2.1 has an immediate effect on the rate of deformation, which increases by a factor of 2.7 (Fig. 4c). We propose that in our experiment, the rate of mineral deformation is almost directly proportional to the applied stress. Note that we only have one piece of data to support this conclusion. Comparison with PSC on halite PSC indenter experiments with ex situ deformation measurement. The experiments with a saturated solution of calcite in 5% solution of NH4C1 show indentation rates of about 1 p~m/ day (Table 1). This value is the same as in the experiments of Gratier (1993), where the rate of halite monocrystal deformation in the same type of experiments with diffusion control was found to be about 1 txm/day at the applied stress of about 50 MPa and with an indenter
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS diameter of 200 txm. Given that the solubility of calcite in an NH4C1 solution is 700 times lower than that of halite, that the rate of calcite indentation should also be slowed by a factor of 2 compared to halite because of the difference in the indenter diameters (300 txm in calcite experiments and 200 Ixm in halite ones), and that this rate should be enhanced several times because of the higher applied stress used in our experiments (the rate of mineral deformation by diffusion-controlled pressure solution is proportional to the applied stress in the case of moderate stress and in exponential proportion to the stress if the stress is high), it can be estimated that calcite deforms two orders of magnitude faster than it should when considering the difference in solubility between calcite and halite.
High-resolution PSC experiments with continuous deformation recording. The observed rate of calcite indentation is of the order of several tens of nm per hour (see Table 3). It should be noted that this rate does not change significantly during the experiment. This is an important difference between our results and those described in Dysthe et at. (2003) for similar experiments with halite single crystals with diffusion control, where the indentation rate varies by several orders of magnitude throughout the experiment: from 2 txm/h at the start, when stress is imposed, to 2 nm/h after several days. However, on most of the calcite indentation curve (from 10 to 100 h) the indentation rate falls in the range between 27 and 180 nm/h (Table 2). These data show that calcite deforms only 5 to 10 times slower than halite, whereas its solubility is 20 000 times lower. For our experiments, the stress may be estimated at 100MPa. This stress is 25 times higher than that used in the halite experiments. This difference enhances the rate of mineral indentation by two orders of magnitude. In addition, the difference in the lengths of diffusion paths (50 Ixm in our experiments compared to 30 Ixm in the halite indentation experiments) slows down this rate by a factor of 2.5. However, even with these explanations, the rate of calcite deformation remains two orders of magnitude higher than it should be if compared with halite when considering the difference in solubility. Comparison between the two types of experiments. If this method is used to compare the results of Gratier (1993) and Dysthe et al. (2002, 2003) for the rate of halite indentation, it would be found that these two rates are of the same order of magnitude provided that they
93
are scaled with respect to the stress and the indenter diameter. On the other hand, for the two types of calcite experiments, the deformation rate is two orders of magnitude higher than it should be when compared with halite considering the difference in solubility. Consequently, there are some additional effects that may be responsible for the fast rates measured on calcite.
Effect of microcracks For an additional explanation, the structure of the crystal surface should be considered. It is known that in diffusion-controlled PSC, the presence of microcracks in the region of dissolution under the indenter dramatically enhances the rate of mineral indentation by a significant reduction in solute transport distances in the trapped fluid phase (Gratz 1991; den Brok 1998; Gratier et al. 1999). The reason for this effect might be related to the substantial difference in transport properties between the trapped fluid phase along the grain contact and the bulk fluid (Rutter 1976). Because of this, the diffusion flux in microcracks (bulk fluid) is greater than along the trapped fluid phase. If the dissolution surface is cut by microcracks, the average diffusion distance in the trapped fluid phase is significantly reduced and the deformation rate is increased dramatically (den Brok 1998). This effect is believed to be critically important in our experiments, as a considerably high stress is applied on a brittle mineral. Cracks seen on the interferometry images of the indenter imprints in the high-resolution PSC experiments provide support for this assumption. The wide scatter of hole depths obtained in the same experimental conditions (Fig. 3) also confirms the microcrack hypothesis. The microcrack structure is unique for each experiment, which is the reason for the significant variation of indentation rates. Precise understanding of this interface structure needs further investigation. Microcracks might be formed by the release of elastic forces when the indenter is left on the crystal. Another effect is the formation of twinning that can be readily achieved in laboratory conditions. Twinning could lead to localized hardening and could promote further fracturation. This is a fast process that potentially happens in the early stages of the deformation process and controls the initial fracture spacing. At this stage, it can only be unequivocally assessed that the calcite interface structure under the indenter is radically different compared to that of halite. This result indicates that great care must be taken in extrapolating the kinetic
94
S. ZUBTSOV ETAL.
data of pressure solution creep obtained from one mineral to another.
Conclusions This article reports on the first experimental study of calcite monocrystal indentations by the pressure-solution creep. 9 The rate of mineral deformation was found to be inversely proportional to the square of the indenter diameter, which means that the process is diffusion-controlled. 9 When comparing the experiments with various calcite solubility (aqueous solution versus solution of NH4CI), the rate of calcite indentation was found to be proportional to its solubility. 9 The rate of calcite deformation was found to be almost proportional to the applied stress. 9 The observed rate of calcite indentation is two orders of magnitude higher than it should be when compared with the halite indentation rate in the same type of experiments, taking into account the difference in solubility of these two minerals. This effect could be explained by the presence of microcracks under the indenter phase. Understanding of this effect needs further studies of the interface structure. Our experimental data suggest that PSC deformation on calcite can be quite fast. This process should occur in nature (basin compaction, fault gouge) with a wide range of rates, depending on stress, fluid composition, and crack density. The presence of microcracks should enhance PSC in active faults after an earthquake as new surfaces are created and new 'indenters' are formed. This could explain the fast fault creep observed after a major earthquake (Donnellan & Lyzenga 1998). The project has been supportedby the ANDRA, the CNRS (Action Thrmatique Innovante), the Norwegian Research Council through the Fluid Rock Interaction Strategic University Program (grant 113354-420), and the Center for Advances at the Academy of Science of Norway. We would like to thank R. Guiguet, D. Tisserand, C. Pequegnat and L. Jenatton for their technical help, and D. Gapais, S. Reddy and B. den Brok for their constructive reviews.
References ALCANTAR, N., ISRAELACHVILI,J. & BOLES, J. 2003. Forces and ionic transport between mica surfaces: Implications for pressure solution. Geochimica and Cosmochimica Acta, 67, 1289-1394.
BAKER, P. A., KASTNER, M., BYERLEE, J. D. & LOCKNER, D. A. 1980. Intergranular pressure solution and hydrothermal recrystallisation of carbonate sediments - an experimental study. Marine Geology, 38, 185-203. DE MEER, S. & SPIERS, C. J. 1997. Uniaxial compaction creep of wet gypsum aggregates. Journal of Geophysical Research, 102, 875-891. DE MEER, S. & SPIERS, C. J. 1999. On mechanisms and kinetics of creep by intergranular pressure solution. In: JAMVEIT, B. & MEAKIN, P. (eds) Growth and Dissolution in Geo-Systems. Kluwer Academic Publishers, Dodrecht, 345-366. DEN BROK, S. W. J. 1998. Effect of microcracking on pressure-solution strain rate: the Gratz grainboundary model. Geology, 26, 915-918. DEN BROK, S. W. J., ZAH1D,M. & PASSCHIER, C. W. 1999. Pressure solution compaction of sodium chlorate and implications for pressure solution in NaCl. Tectonophysics, 307, 297-312. DONNELLAN, A. & LYZENGA, D. A. 1998. GPS measurements of fault afterslip and upper crustal deformation following the Northridge earthquake. Journal of Geophysical Research, 103, 21 28521 297. DYSTHE, D. K., PODLADCHIKOV, Y., RENARD, F., FEDER, J. G. & JAMTVEIT, B. 2002. Universal scaling in transient creep. Physical Review Letters, 89, 246 102-1-246 102-4. DYSTHE, D. K., RENARD, F., FEDER, J. G., JAMTVEIT, B., MEAKIN, P. & JOSSANG, T. 2003. High resolution measurements of pressure solution creep. Physical Review E, 68, 011603. GRATIER, J. P. & GUIGUET, R. 1986. Experimental pressure solution-deposition on quartz grains: the crucial effect of the nature of the fluid. Journal of Structural Geology, 8, 845-856. GRATIER, J. P. 1993. Experimental pressure solution of halite by an indenter technique. Geophysical Research Letters, 20, 1647-1650. GRATIER, J. P., RENARD, F. & LABAUME, P. 1999. How pressure solution creep and fracturing processes interact in the upper crust to make it behave in both brittle and viscous manner. Journal of Structural Geology, 21, 1189-1197. GRATZ, A. J. 1991. Solution-transfer compaction of quartzites: Progress towards a rate law. Geology, 19, 901-904. JONES, R. V. 1988. Instruments and Experience. Wiley, New York. LAFAUCHEUX, F. 1974. Contribution h l'dtude de ddfauts pMsent~s par des calcites hydrothermales de synthkse. Thrse d'Etat, Paris VI. MORSE, J. W. & ARVIDSON,R. S. 2002. The dissolution kinetics of major sedimentary carbonate minerals. Earth-Science Reviews, 58, 51-84. PATERSON, M. S. 1973. Nonhydrostatic thermodynamics and its geological applications. Reviews in Geophysics, 11, 355-389. PHARR, G. M. & ASHBY, M. F. 1983. On creep enhanced by a liquid phase. Acta Metallurgica, 31, 129-138. PLUMMER, L. N., WIGLEY, T. M. L. & PARKHURST, D. L. 1978. The kinetics of calcite
PRESSURE SOLUTION ON CALCITE MONOCRYSTALS dissolution in CO2-water systems at 5 ~ to 60 ~ and 0.0 to 1.0 atm C O 2. American Journal of Science, 278, 179-216. RAJ, R. 1982. Creep in polycrystalline aggregate by matter transfer through a liquid phase. Journal of Geophysical Research, 87, 4731-4739. RAMSAY,J. R. 1967. Folding and Fracturing of Rocks. McGraw-Hill, London. REDDY, M. M. & WANG, K. K. 1980. Crystallization of calcium carbonate in the presence of metal ions. I. Inhibition of magnesium ion at pH 8.8 and 25 ~ Journal of Crystal Growth, 50, 470-480. RUTTER, E. H. 1976. The kinetics of rock deformation by pressure solution. Philosophical Transactions of the Royal Society of London, 283, 203-219. RUTTER, E. H. 1983. Pressure solution in nature, theory and experiment. Journal of the Geophysical Society of London, 140, 725-740.
95
SKVORTSOVA,Z. N., TRASKIN,V. Y., LOPATINA,L. I. & PERTSOV, N. V. 1994. Adsorption-induced decrease of the recrystallisation creep rate. Colloid Journal, 56, 177-179. TADA,T. & SIEVER,R. 1986. Experimental knife-edge intergranular pressure solution of halite. Geochimica and Cosmochimica Acta, 50, 29-36. WEYL, P. K. 1959. Intergranular pressure solution and force of crystallization - a phenomenological theory. Journal of Geophysical Research, 64, 2001-2025. ZANG, X., SALEMANS,J., PEACH, C. J. • SPIERS, C. J. 2002. Compaction experiments on wet calcite powder at room temperature: evidence for operation of intergranular pressure solution. In: DE MEER, S., DRURY, M. R., DE BRESSER, J. H. P. PENNOCK, G. M. (eds) Deformation Mecha-
nisms, Rheology and Tectonics: Current States and Future Perspectives. Geological Society, London, Special Publications, 200, 29-39.
Contrasting microstructures and deformation mechanisms in metagabbro mylonites contemporaneously deformed under different temperatures (c. 650 ~ and c. 750 ~ L. B A R A T O U X 1'2'5, K. S C H U L M A N N 3, S. U L R I C H t'4 & O. L E X A t
~Institute of Petrology and Structural Geology, Charles University, Albertov 6, 12843, Prague, Czech Republic (e-mail: Ika @natur.cuni.cz) 2UMR 5570, ENS and Lyon 1 University, 2 rue RaphaYl Dubois, 69622, Villeurbanne Cedex, France 3Universitd Louis Pasteur, EOST, UMR 7517, 1 Rue Blessig, Strasbourg, 67084, France 4Geophysical Institute, Czech Academy of Sciences, Bo(n{ 11/1401, 14131 Praha 4, Czech Republic 5Present address." Czech Geological Survey, Klarov 3, Praha 1, 11821, Czech Republic Abstract: Deformation mechanisms of amphibole and plagioclase were investigated in two metagabbroic sheets (the eastern and western metagabbros from the Stars M~sto belt, eastern Bohemian Massif), using petrology, quantitative microstructural and electron backscattered diffraction methods. After the gabbroic pyroxene was replaced by amphibole, both gabbroic bodies became progressively deformed. The eastern metagabbros were deformed under temperature of c. 650 ~ and the western metagabbros under c. 750 ~ Subgrain rotation and dislocation creep, characterized by strong crystallographic and shape preferred orientations, operated in plagioclase of the eastern belt during the early stages of deformation. Subsequent randomizing of plagioclase crystallographic preferred orientation is interpreted to be due to grain boundary sliding in the mylonitic stage. Large (50150 ixm) grain sizes during the mylonitic stages are interpreted to be due to low strain rates. Amphibole is stronger and deforms cataclastically, leading to important grain size reduction when the bulk rock strength drops substantially. In the western belt, plagioclase deformed by dislocation creep accompanied by grain boundary migration (possibly chemically induced) while heterogeneous nucleation and syndeformational grain growth in conjunction with dislocation creep were typical for amphiboles.
Microstructural and rheological behaviour of natural polyphase rocks is a complex problem, which has been studied in detail mostly in quartzo-feldpathic rocks (e.g. Gapais 1989; Handy 1990). In these rocks, feldspars are generally considered as strong minerals while quartz represents the weak phase. Handy (1994) proposed a scheme for rocks containing minerals of contrasting rheology with two end-members: load-beating framework (LBF) and interconnected weak layers (IWL), based on the assumption that at least one mineral (generally the weaker one) is deformed by the mechanism of dislocation creep. As pointed out by Jordan (1988) the LBF is not stable with increasing strain, resulting in mechanically induced compositional layering. The strength
of a two-phase material depends on the proportion of the weak mineral and on the rheological contrast between the two phases (Jordan 1988; Handy 1990, 1994). However, if the deformation mechanism switches from crystal plasticity to some grain size sensitive process, the strength of the bulk rock drops rapidly with respect to the preceding model due to strain softening (Knipe 1989). It was also shown that generally stronger feldspars may become even weaker than quartz if they deform by diffusional creep (e.g. Voll 1976; Simpson 1985; Gapais 1989; Martelat et al. 1999). Equivalent comparative microstructural study of plagioclase and hornblende in naturally deformed metabasic rocks as well as experimental studies of the rheology of amphibole-plagioclase
From: GAPA1S,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. DeformationMechanisms,Rheologyand Tectonics:from Mineralsto the Lithosphere. Geological Society, London, Special Publications, 243, 97-125. 0305-8719/05/$15.00 ~') The Geological Society of London 2005.
98
L. BARATOUX E T A L .
bearing rocks are rare (e.g. Hacker & Christie 1990; Wilks & Carter 1990). However, metabasites are considered to constitute a major part of the lower continental crust (Rutter & Brodie 1992) and the study of deformation microstructures and mechanisms is therefore crucial for understanding the deformation behaviour and rheology of mafic tectonites. Plagioclase rheology is believed to control the strength of the lower crust in several models (e.g. Carter & Tsenn 1987; O r d & Hobbs 1989). Hornblende is assumed to be a relatively strong phase and to behave passively during deformation unless it forms a load-supporting framework (Brodie & Rutter 1985). Many studies concerned with deformation mechanisms of plagioclase and hornblende have been published to date. Experiments suggest that plagioclase is deformed by dislocation creep under lower crustal conditions (e.g. Tullis & Yund 1987; Ji & Mainprice 1990; Kruse et al. 2001) with commonly active (010)[001] principal slip system (e.g. Olsen & Kohlstedt 1985; Kruhl 1987). Kruhl (1987) also observed a (001) [ 100] slip system in naturally deformed plagioclases and Stiinitz et al. (2003) have shown that slip on (001) and {111} in (110) direction are similarly active in experimentally deformed An60 crystals. Besides these examples, other less common slip systems were suggested by Marshall & McLaren (1977a, b), Montardi & Mainprice (1987), and Olsen & Kohlstedt (1984, 1985). In addition, grain size sensitive processes in plagioclases have been referred to in some studies (Boullier & Gu6guen 1975; Lapworth et al. 2002). Hornblende generally undergoes brittle deformation under low temperature conditions (e.g. Brodie & Rutter 1985; Nyman et al. 1992; Lafrance & Vernon 1993). Many studies document crystal plastic deformation of hornblende with a dominant (100) [001] slip system from experiments and natural rocks (e.g. Rooney et al. 1970; Cumbest et al. 1989a, b; Hacker & Christie 1990). Volume diffusion rates are, however, slow in amphiboles (e.g. Freer 1981) implying that dislocation climb is limited even at geological strain rates. Syndeformational chemical reactions between hornblende and plagioclase are common in metabasic rocks (Brodie 1981; Brodie & Rutter 1985) and involve deformation mechanisms such as chemically induced grain boundary migration (CIGM) (Cumbest et al. 1989a), on nucleation of new plagioclase (Rosenberg & Stfinitz 2003) or hornblende at plagioclase grain boundaries (Kruse & Sttinitz 1999). Metagabbros from the Star6 M6sto belt (eastern margin of the Bohemian Massif) represent
an example of a two-phase metabasic system deformed at different temperatures and strain gradients. The aim of this study is to show two types of progressive evolution of deformation microstructures and textures of dynamically recrystallized plagioclase-hornblende bearing metagabbros at amphibolite and upper amphibolite facies conditions with increasing bulk strain. Methods such as quantitative textural analysis, crystallographic preferred orientations (CPO) and study of mineral chemistry were used to constrain deformation mechanisms for both minerals. Finally, deformation mechanisms of rheologically contrasting plagioclase and hornblende are correlated and the mechanical behaviour of mafic rocks deformed under lower crustal metamorphic conditions is discussed.
Geological setting The Star6 M6sto (SM) belt in the eastern margin of the Bohemian Massif separates the high grade gneisses of thickened continental crust of the Lugian domain in the west from a NeoProterozoic continental margin in the east (Fig. l a). The SM domain was thinned during Cambro-Ordovician rifting and underwent granulite facies metamorphism (Stfpskfi et al. 2001). Subsequent Variscan tectonics resulted in N E SW trending structures dipping at relatively high angles to the west (Figs 1a and b). Strongly deformed 'western' metagabbros of Cambro-Ordovician protolith ages occur at the top of the SM belt (Fig. 1) (Kr6ner et aL 2000), The metagabbros were pervasively affected by a ductile shear zone along which was emplaced syntectonically a Carboniferous tonalite sill dated at 340 Ma (Stfpskfi et aL 2004). A carboniferous metamorphism of adjacent metagabbros reached higher amphibolite facies conditions because of the strong heat input from the tonalite sill. A leptyno-amphibolite complex of CambroOrdovician age comprising a sequence of alternating amphibolites and tonalitic gneisses occurs in the footwall of the tonalite sill. This complex is underlined by the 'eastern' metagabbro sheet, which is supposed to form part of the same lower crust as the western (upper) metagabbro sheet (Stfpskfi et aL 2001). Amphibolite facies Carboniferous metamorphic conditions of the eastern (lower) metagabbro are also attributed to the heat input of a more distant hangingwall tonalite intrusion. Rocks of Cambro-Ordovician age suffered two deformations: D~ of Cambro-Ordovician age and D2 of Carboniferous age. The latter one prevails in most lithologies of the SM belt, marked by a penetrative west-dipping $2 foliation bearing a
TEXTURES OF NATURALLY DEFORMED METAGABBROS
99
Fig. 1. (a) Location of the studied area in the frame of the European Variscides. Geological map of the Star6 M~sto belt is based on unpublished geological maps at 1:25 000 provided by courtesy of the Czech Geological Survey (Dr. M. Opletal, author). Important thrusts and faults as well as location of samples and a cross-section A-A' (Fig. lb) are indicated. (b) Schematic cross-section based on field observations shows major structures, lithology and major tectonic boundaries. (Vertical axis not to scale.) subhorizontal or gently inclined N - S trending mineral lineation. The eastern metagabbros in the hangingwall of the high-grade rocks of the Silesian d o m a i n were affected by localized D2 shear zones that developed under amphibolite
facies conditions attained during the peak of the Carboniferous m e t a m o r p h i s m . Identical geometry of structures and kinematics suggest that the eastern and western metagabbro belts were d e f o r m e d under a dextral transpressive shear
100
L. BARATOUX ETAL.
regime but under different thermal conditions (Stfpsk~i et al. 2001).
Methods and techniques Mineral chemistry
Minerals were analysed with a Cameca SX 100 electron microprobe equipped with four WDS spectrometers at Blaize Pascal University, Clermont-Ferrand, France. Operating conditions were 15kV, 10hA beam current, 2-5p~m beam diameter, 20 s counting time, and natural mineral standards. Some hornblende and plagioclase were analysed using a CamScan $4 scanning electron microscope and attached Link EDX microanalytical system, at Charles University, Prague, Czech Republic. Operating conditions were 20 kV, 1.8 nA beam current, 1-3 p,m beam diameter, 120 s counting time, and mineral standards Structure Probe Instruments (SPI). Ca maps from plagioclase were made at Claude Bernard University in Lyon, with operating conditions 15 kV, 15 nA beam current and spatial resolution of 512 x 512 pixels. Quantitative microstructural analysis
Quantitative microstructural analysis of grain boundaries was carried out on traced and digitized outlines of grains in ESRI ArcView 3.2 Desktop GIS environment. The map of grain boundaries was generated using ArcView extension Poly (Lexa 2003). The resulting polygons have been treated by MATLAB TM PolyLX Toolbox (http://petrol.natur.cuni.cz/~ ondro; Lexa 2003), in which grain shape and grain boundary preferred orientations (SPO and GBPO, respectively) were analysed using the moments of inertia ellipse fitting and eigenanalysis of bulk orientation tensor techniques (Lexa, 2003; modified SURFOR technique by Panozzo (1983) for GBPO). Their degree is expressed as the eigenvalue ratio (r = E1/E2) of the weighted orientation tensor of grain shapes or boundaries. The orientation is defined by V~ and V2 eigenvectors. The grain sizes of the minerals were calculated in terms of Ferret diameters of grain section without stereological corrections. Crystallographic preferred orientation
Amphibole and plagioclase crystallographic preferred orientations (CPO) were collected using a scanning electron microscope CamScan $4 in Prague and a JEOL JSM 5600 in Montpellier equipped with Channel5 electron backscatter
diffraction (EBSD) system from HKL Technology (Prior et al. 1999). Thin sections were polished using 0.25 ~m diamond paste. To remove all surface damage and minimize relief between minerals, sections were chemically polished using a colloidal silica suspension. All thin sections were carbon-coated. The coating reduces the quality of the electron backscatter diffraction patterns (EBSP) so that automatic indexing mode of the EBSP system could not be used. Most data were therefore collected manually. Operating conditions were 20 kV in Prague and 15 kV in Montpellier, 5.6 nA beam current, working distance 39 ram, and 2 - 5 ~m beam diameter. For each measurement, three Euler angles (vl, 05, v2) characterizing the lattice orientation as well as the nature of the mineral were determined and stored. Practice shows that plagioclase diffraction patterns do not change significantly from albite up to at least An65 (Lapworth et al. 2002). Therefore, the Anorthite48 database was used for indexing plagioclase. Pole figures and inverse pole figures were projected using the software developed by D. Mainprice (ftp://ftp.dstu.univmontp2.fr/pub/TPHY/david/pc). Projections of crystallographic axes ([a], [b] and [c]) or optical indicatrix (o~, /3, and 3/) are generally used for plagioclase (e.g. Jensen & Starkey 1985; Olsen & Kohlstedt 1985; Ji & Mainprice 1988, 1990; Prior & Wheeler 1999). The degree of CPO has been quantified using orientation tensor of crystallographic planes and directions (Mainprice, ftp://ftp.dstu.univmontp2.fr/pub.TPHY/david/pc). The intensity of CPO is given by the I parameter (Lisle 1985): 3
I = 15/2 x Z ( E i
- 1/3) 2
i=1
where E~, E2, and E3 represent the eigenvalues of the preferred orientation of poles to planes and directions of amphibole and plagioclase grains plotted in pole figures. The values of I range between 0 (no preferred orientation) and 5 (all fabric elements perfectly parallel one to another).
Microstructures The metagabbros from the eastern (lower) belt are affected by localized shear zones (Fig. lb) and all stages from non-deformed rock, protomylonite, mylonite, and up to highly strained ultramylonite are present (Figs 2a, b, c and d). In the western metagabbro (upper) belt, the protolith stage is not preserved and only two deformational stages can be distinguished
TEXTURES OF NATURALLY DEFORMED METAGABBROS
101
Eastern belt (-650 ~
Western belt (-750 ~
Fig. 2. Macro photographs of typical metagabbro structures from the eastern (a-d) and the western (e, t3 belts: (a) non-deformed metagabbro (E0); (b) protomylonite (El); (c) mylonite (E2); (d) ultramylonite (E3); (e) augen mylonite (W1); (f) banded mylonite (W2).
(Figs 2e and f): mylonites with hornblende porphyroclasts related most probably to the Cambro-Ordovician metamorphic event and completely recrystallized mylonites characterized by a monomineral layering.
Metagabbros of the eastern belt consist of plagioclase (40-60%), amphibole (40-60%), relict pyroxene in the cores of some large amphibole grains (less than 1%), and titanite (less than 1%). No substantial change in modal proportions
102
L. BARATOUX ETAL.
is observed with increasing deformation. Modal proportions of amphibole and plagioclase in the western metagabbros are more variable than those in the eastern belt. Amphibole proportion may vary between 20 and 80%, most likely due to original magmatic compositional variations. For the purpose of our study, samples composed of c. 50% of amphibole and c. 50% of plagioclase were chosen.
Deformation of the eastern (lower) metagabbro sheet Non-deformed metagabbro (EO). At low strain, plagioclase and hornblende exhibit euhedral randomly oriented crystals of 0.5-3 mm in size. Tapering mechanical twins and local undulatory extinction occur in plagioclase. There is no evidence for any kind of dynamic recrystallization or crystallization of new grains. Undulatory extinction locally attests to some bending of amphibole grains. Amphibole porphyroclasts show random spatial distribution and they are generally not interconnected.
Protomylonite (El). At higher strains, about 20-25% of the total volume of plagioclase grains but only 8-10% of amphibole show
evidence of strain, suggesting that the deformation was accommodated mostly by plagioclase recrystallization and associated grain-size reduction. Plagioclase grains show polysynthetic twins according to albite and pericline laws (Tullis 1983) and patchy undulatory extinction. Large plagioclase porphyroclasts of 2 - 5 mm are cut by brittle fractures (Fig. 3a) reducing the grain size to 0.5-1 mm. The fractures are filled with small twin-free recrystallized grains of 0.02-0.1 mm. Two recrystallization mechanisms have been identified: bulging and subgrain rotation recrystallization (Fig. 3b), as proposed by Poirier & Guillop6 (1979) or Fitz Gerald & StiJnitz (1993), leading to core-mantle structures. Core and mantle structures were observed with an intermediate zone of subgrains and new grains of similar size developed along porphyroclast boundaries. Boundaries between the neoblasts become progressively straight, meeting at triple junctions of 120 ~. Large porphyroclasts of hornblende reveal strong internal deformation such as kinking, bending leading to sweeping undulatory extinction and (100) twinning. The twin planes are not regular and they locally form finger-like structures. Brittle fractures transecting large grains are often present. Randomly oriented porphyroclasts of 2 - 5 m m in size rotate
Fig. 3. Drawing and micrographs(XPL) of the eastern metagabbro protomylonite. (a) Initial stage of plagioclase deformation characterized by brittle fractures cross-cutting large porphyroclasts. Digitizeddrawing was used for the quantitative textural analysis.Arrows mark three clasts derived by fracturing of one porphyroclast and show corresponding grains in the digitized drawing and micrograph. Plagioclaseis white, hornblendeis light grey, and opaque minerals are black. (b) New grainsdevelop preferentiallyalong fractures by a mechanismof subgrainrotation (SR). Note that the neoblasts are twin-free. (P) refers to porphyroclast.
TEXTURES OF NATURALLY DEFORMED METAGABBROS progressively their cleavage planes {110} parallel to the foliation. Small needle-like grains (0.020.05 x 0.1-0.3 mm in size), arranged parallel to the mylonitic foliation, are present in highstrain domains.
Mylonite (E2). Within the eastern metagabbros, plagioclase porphyroclasts are recrystallized into elongate aggregates or bands wrapped around hornblende clasts, and porphyroclasts represent only 5 - 1 0 % of the total plagioclase volume. Elongate needle-like hornblende grains are
103
arranged into aggregates or bands parallel to the main mylonitic foliation (Fig. 4a). Isolated hornblende needles scattered within plagioclase-rich domains are inclined at an angle of 2 0 - 3 0 ~ with respect to the main fabric, indicating noncoaxial deformation (Figs 4a and b). Plagioclase grains attain 0.1-0.25 mm in size. Increase in grain size with respect to the protomylonite stage and common optical zoning of recrystallized grains due to increase in anorthite content suggest syndeformational growth. Matrix grains are often twin-free; the twinned grains represent
Fig. 4. Digitized drawings and micrographs of the eastern mylonite. (a) Mylonitic foliation with alternating layers of amphibole and plagioclase aggregates (PPL). (b) Plagioclase of subequant shapes and grain size typical for the mylonitic stage (E2-pll). (c) High-strain plagioclase domain (E2-p12) adjacent to an amphibole porphyroclast is characterized by elongate shapes and strong SPO (XPL). (d) Amphibole porphyroclast (E2) cross-cut by microshear zones filled by small needle-like grains (XPL). Undulatory extinction documents strong internal deformation. Arrows mark corresponding grains in the digitized drawings and micrographs.
104
L. BARATOUX ETAL.
only 10-20% of the total plagioclase volume. Plagioclases generally exhibit subequant shapes with straight grain boundaries meeting at 120 ~ triple joints (Fig 4b). However, in the vicinity of amphibole porphyroclasts, the plagioclase grains are more elongate with higher aspect ratios up to 2.5 (Fig. 4c). Hornblende grains are arranged in anastomosing interconnected bands defining the mylonitic foliation. Most of the hornblende matrix grains, 0.1-0.3 mm in length, are elongate in XZ sections attaining high aspect ratios (up to 6) while they are lozenge-shaped in the YZ sections with aspect ratios of 1.5-3. A total of 10-20% of the hornblende is present in a form of locked-up porphyroclasts of 0.2-2 mm in size showing undulatory or patchy extinction characteristic of strong internal strain. These porphyroclasts are elongate parallel to the foliation. Porphyroclasts are often transected by microshear zones (Fig. 4d) and small needle-shaped grains consequently develop within these zones by activation of the weak cleavage system {110}.
Ultramylonite (E3). The ultramylonitic metagabbros of the eastern sheet are marked by very uniform mineral banding of fine-grained hornblende and plagioclase layers ranging between 1 and 5 mm in width (Fig. 5a). Asymmetric fabric of hornblende and plagioclase is no longer present. Plagioclase forms continuous bands including abundant interstitial grains of amphiboles
(Fig. 5b). In contrast, amphibole-rich bands are almost monomineral. Plagioclase is completely recrystallized, varying in size between 0.05 and 0.3mm. The grains have smooth rounded shapes locally elongate parallel to the foliation. Rounded grains of 0.5 mm in size, interpreted as relicts of original porphyroclasts, locally occur in the fine-grained matrix. Plagioclase grains are affected by strong retrograde sericitization and albitization. Interstitial and matrix grains of hornblende reach 0.03-0.1 x 0.1-0.3 mm in size. Lockedup porphyroclasts of amphibole (0.5-1 mm in size) occur parallel to the foliation. However, they are less common than in the mylonite (E2). All amphibole grains are characterized by very strong SPO (Fig. 5c). The mylonitic foliation is cross-cut by extensional veins filled by a mixture of epidote and amphibole. These veins show that fluid activity was high after the main mylonitic episode. Albitization and sericitization of plagioclase in some samples suggest that increased amount of water was present in these rocks also during late retrogression, which is most likely not related to the main process of mylonitization.
Deformation of the western (upper) metagabbro sheet Augen mylonite (W1). In the western metagabbro belt, the deformation was pervasive so that initial protomylonite stages of deformation
Fig. 5. Drawingsand micrographof the eastern ultramylonite.(a) Plagioclaseand amphibolebanding (PPL). Plagioclase is darker than amphiboledue to strong sericitization.(b) Plagioclaseband in ultramylonite(E3) with abundant interstitialhornblende. Small grain size and subequant shapes are typical for these plagioclases. (c) Amphibolelayer from ultramylonite(E3) is characterizedby very fine grain size and strong SPO.
TEXTURES OF NATURALLY DEFORMED METAGABBROS are absent. The least deformed rock represented by augen mylonite is marked by the presence of rare 2 - 3 mm large porphyroclasts of plagioclase (Fig. 6a) and by 1 - 2 r a m wide and 4 - 5 mm long porphyroclasts of hornblende within the recrystallized matrix (Fig. 6b). This rock type is preserved only in the southern part of the metagabbro belt. Plagioclase porphyroclasts represent only 5-10% of the total plagioclase volume and show deformational twins and undulatory extinction. Recrystallized plagioclase grains show optical zoning. Modal proportion of amphibole porphyroclasts cannot be estimated as they differ from the neoblasts only in their chemical composition, making them optically indiscernable. Amphibole porphyroclasts exhibit patchy or sweeping undulatory extinction but deformation twins have not been observed. The local occurrence of twins is assumed to be growth-related, the twin planes being always perfectly straight. Prismatic porphyroclasts attain higher aspect ratios than more rounded neoblasts. All rocks in the western belt are affected by amphibolite facies re-equilibration partly destroying the granulite facies peak metamorphic assemblages.
105
Banded mylonite (W2). Distinct monomineral bands of 1-10 mm in width are typical of these highly deformed mylonites. Small hornblende grains of irregular shape are common within the plagioclase layers, but plagioclase is rarely included within amphibole bands. Both minerals are completely recrystallized. Hornblende is in some places replaced by later prisrnatic cummingtonite, which grows either parallel or obliquely to the mylonitic foliation and is locally included within the plagioclase layers. Two types of plagioclase microstructures occur within one plagioclase band. The first type is characterized by grain sizes ranging from 0.3 to 1 mm in diameter, subequant or slightly elongate, but always irregular, grain shapes with serrated boundaries (Fig. 7a), and growth-related optical zoning. Mechanical twins cross-cut the optical zoning in some cases (Fig. 7b), suggesting that at least some of them formed later. The second type is characterized by narrow (c. 0.5 mm) high-strain zones trending parallel or slightly oblique to the layering (Fig. 7c). Here, the grain size is reduced to 0.05-0.2 ram. Plagioclase grains attain higher aspect ratios (up to 2) and exhibit undulatory extinction and subgrain boundaries elongate parallel to the high-strain zone. The contact between these two deformational domains is either sharp or transitional. Hornblendes of 0.1 - 1 mm in size have straight grain boundaries locally meeting at high angles suggesting a high degree of textural equilibration (Vernon 1976) (Fig. 7d). No undulatory extinction occurs and twinning is very rare. Where present, the twin planes are always very straight, indicating that twinning is growth-related rather than deformational. Aspect ratios vary between 1.5 and 2.5, but some grains with aspect ratio close to 1 occur. Grains with high aspect ratios are not always parallel to the mylonitic foliation.
Bulk rock chemistry, mineral chemistry and zoning Bulk rock chemistry
Fig. 6. Drawingsand micrographsof the western augen mylonite. (a) Digitizeddrawing of plagioclase(W1) used for the quantitative textural analysis.Large porphyroclasts are surrounded by a mantle of finegrained matrix grains. (b) Hornblende(W1) shows variable grain size and strong SPO. Plagioclaseis white, hornblende is light grey, and opaque mineralsare black.
Table 1 presents bulk rock chemical analyses studied by X-ray fluorescence at Claude Bernard University, Lyon, France. The eastern and western metagabbro sheets are regarded to be a part of the same lower crustal unit prior to their involvement into the Variscan tectonics (Stfpskfi et al. 2001). However, their chemistry is not identical and bulk rock analyses (Table 1) show that western metagabbros are characterized by higher contents in A1 and Ca and lower contents in Na and Fe than their
106
L. BARATOUX ET AL.
Fig. 7. Drawings and micrographs of the western banded mylonite. (a) Lobate boundaries document grain boundary migration (GBM) in coarse-grained plagioclases of banded mylonite (XPL). (b) Mechanical twins (MT) transect the growth-related zoning (GZ) in banded mylonites documenting a later deformation phase (XPL). (c) Digitized drawing of two plagioclase domains (W2-pll and W2-p12) used for quantitative textural analysis. (d) Hornblendes (W2) are characterized by straight grain boundaries meeting occasionally at triple points at variable angles. Many small grains indicating high nucleation rate are present. SPO is rather low. Plagioclase is white, hornblende is light grey, pyroxene is dark grey, and opaque minerals are black in all drawings.
eastern counterpart. The increase in Na and depletion in Ca in both metagabbro belts is related either to mylonitization or to primary variations due to postmagmatic processes. Trace elements show very similar trends for the eastern and western metagabbros, suggesting that they may originate from the same source. Metagabbros from both belts display a slightly negative Nb anomaly normalized to MORB (Sun & McDonough 1989). Augen mylonites from the western belt are depleted in Zr and Ti compared to eastern mylonites. Contents of Rb, Ba, and K are strongly variable in both belts, which is most probably related to postmagmatic alterations and/or mobility of these elements during metamorphism.
Mineral chemistry and zoning Syndeformational chemical reactions are common in metabasic rocks (Brodie 1981; Brodie & Rutter 1985). Variations in plagioclase composition are shown in Figures 8 and 9. Amphibole compositions according to the classification of Leake et al. (1997) are plotted in Figure 10. Representative analyses of mineral compositions are listed in Tables 2 and 3.
Mineral abbreviations are according to Kretz (1983). It is likely that the studied metagabbros experienced a metamorphic event prior to the main deformation and metamorphism studied in this work. This is supported by existence of incompletely amphibolized pyroxenes in undeformed rocks of low-strain domains. In addition, the compositions of cores of large amphibole porphyroclasts (actinolites to actinolitic magnesiohornblendes) deviate from those of recrystallized magnesio-hornblendes in the eastern belt. Such compositions of primary amphiboles may indicate early greenschist to amphibolite facies conditions preceding the main higher grade Carboniferous deformation. The peak metamorphic assemblages within the eastern metagabbro consist of P1 + Hbl + Qtz + Ttn _+ Ilm __%Mag. The composition of plagioclase porphyroclasts in protomylonite (El) varies between Anso and An6o (Fig. 8a). Small recrystallized grains show a composition similar to the mother host grain corresponding to Anso-6o. More albitic compositions (An4o_45) occur along grain boundaries or triple junctions of recrystallized grains (Fig. 9a). The geometry and sharp gradient in mineral zoning crosscutting several grains is attributed to post-peak
TEXTURES OF NATURALLY DEFORMED METAGABBROS
107
Table 1. Representative analyses of bulk rock compositions Eastern belt
Western belt
Protolith (E0) Mylonite (El) Ultramyl. (E3) Augen myl. (Wl) Banded myl. (W2) Banded myl. (W2) SiO2 TiO2 A1203 FeO T MnO MgO CaO Na20 K20 P205 LOI H20Total
48.61 0.71 17.76 6.14 0.12 9.31 12.47 2.64 0.16 0.04 0.88 0.06 98.90
49.20 0.82 19.74 6.32 0.10 6.56 9.22 3.29 1.52 0.03 1.91 0.12 98.83
52.69 0.94 18.04 7.33 0.13 5.81 8.60 4.99 0.10 0.03 0.57 0.05 99.28
46.23 0.10 21.57 4.86 0.10 7.91 13.15 0.82 1.60 0.01 2.76 0.19 99.30
50.20 0.35 21.43 4.10 0.08 6.20 12.08 2.98 0.51 0.02 1.21 0.09 99.25
48.92 0.37 24.20 3.84 0.06 4.76 11.58 3.55 0.25 0.07 1.65 0.10 99.35
Ba Rb Sr Zr Nb Y V Cr Ni CO Sc Cu Pb Ga XMg*
32.1 1.6 409.4 33.0 1.3 12.2 185.9 550.5 122.8 36.8 42.3 8.0 0.8 13.8 0.730
149.0 55.1 742.0 34.7 1.3 9.4 160.1 214.7 60.6 35.4 24.9 3.6 4.5 15.0 0.649
52.0 4.9 226.9 39.3 1.7 20.2 203.9 222.4 58.8 29.2 30.4 19.1 2.5 12.2 0.586
374.3 74.7 219.1 7.8 0.6 3.3 115.0 136.0 48.9 32.0 41.3 2.9 6.0 12.8 0.744
52.3 26.9 346.3 10.7 1.3 7.2 131.9 622.9 87.3 26.6 33.6 20.4 2.4 16.9 0.729
54.6 11.9 397.4 29.6 1.2 6.4 92.0 269.9 90.9 25.4 16.8 33.3 2.3 17.1 0.688
Major elementsare given in wt%, trace elementsare given in ppm. *XMg= Mg/(Fe2+ + Mg).
a) 12-
Porphyroclasts
E a s t e r n belt
b)
Western
Porphyroclasts
belt
trogression
g o8
~_ 6
i
IIIN!I I].
.
.
I] ,. I
I
,I
Recrystallizedgrains I Late retrogression
2HI 10
20
30
40
50
% An
i
i
i
,
i
,
i
i
III
,
IIII i
i
,
i
i
~ protomylonite(El) Recrystallizedgrains ~[Z] mylonite(E2) r~ ultramylonite(E3) I
~I---1 Banded Augenmylonite rnylonite(W1)1 (W2)l
I,
111 t h
60
70
80
90
10
20
30
40
50
% An
60
70
80
90
Fig. 8. Composition of plagioclase in (a) the eastern and (b) the western belts. Porphyroclasts are depicted separately from matrix grains. Deformation grade increases from dark grey to white colour.
108
L. BARATOUX ET AL.
Fig. 9. BSE images of plagioclaseand amphibole and representative chemical maps of plagioclase compositional variations. (a) Plagioclasein protomylonite from the eastern belt (El) suffersfluidrelated chemical variationsmodifyingthe composition at triple points and grain boundaries. (b) Growth-related chemical zoning in mylonitefrom the eastern belt (E2). (c) Growth-related zoning in coarse-grainedplagioclase of banded mylonitefrom the western belt (W2).
fluid-phase activity along grain boundaries (McCaig & Knipe 1990). Recrystallized matrix plagioclases in the mylonite (E2) have andesitic compositions (An3o-5o) and show increases in anorthite content towards their rims (Fig. 9b), which is commonly ascribed to increasing metamorphic conditions. Rare rounded porphyroclast, as well as matrix grains in ultramylonites (E3), underwent strong and irregularly developed late albitization. Matrix grains correspond to An3o_43 although oligoclase and albite compositions (An6_ 15) are also common along fractures, cleavages and grain boundaries. These irregular compositional variations are attributed to late retrogression unrelated to the main deformation process. Amphibole porphyroclasts in protomylonites (El) are magnesio-hornblendes with variable content in Si but constant XMg (Fig. 10a).
Amphiboles with more actinolitic composition replacing former magmatic pyroxenes can be found in the cores of some grains. Small amphibole grains show similar compositions compared to the porphyroclasts, some of them being more tschermakitic. The amphiboles in mylonites (E2) are magnesio-hornblendes with slightly lower Si contents than those in the protomylonites. High Si/low A1 domains documenting incomplete transformation of pyroxenes into hornblendes are present in the cores of large porphyroclasts as well. Locked-up grains in the ultramylonites (E3) vary between magnesiohornblende and tschermakitic composition. Small new grains have slightly lower A1 contents compared to the less deformed stages. In the western belt, the peak metamorphic assemblages correspond to P1 § Hbl _ Cpx _+ Opx +_ Grt ___Ttn + Ilm. A granulite facies mineral assemblage, comprising also sapphirine and corundum, is not stable and re-equilibrated during subsequent upper amphibolite facies reworking. In the south, granulite facies assemblages are absent and probably completely re-equilibrated. In the north, a late metamorphic phase is documented by the presence of prismatic orthoamphibole (gedrite). Plagioclase porphyroclast compositions are very similar to those of the eastern belt and vary between An4o and An65. Recrystallized matrix grains in the augen mylonite (W1) show locally increasing anorthite content up to An6o_9o compared to the porphyroclasts, indicating that chemical reactions took place during recrystallization. Sharp limited domains of An4o_45 composition along grain boundaries are also common, suggesting late fluid activity (Knipe & McCaig 1994). Recrystallized plagioclase grains in banded mylonites (W2) show stronger variability in compositions (An35-9o). Growth-related zoning due to increasing metamorphic grade is documented in Figure 9c. The cores have more albitic composition (An34) compared to the rims (An6o_65) and are attributed to the early metamorphic stage. Modifications of plagioclase composition characteristic of fluid-related compositional changes are also developed in the banded mylonite. Amphibole porphyroclast compositions (Fig. 10b) were analysed in two different thin sections of augen mylonite (W1) and the difference in their mineral chemistry may be explained by local bulk rock chemical variations. The amphibole porphyroclasts correspond to magnesiohornblendes with similar compositions to those in the eastern belt. Porphyroclast rim compositions show a decrease in Si content with respect to the cores. Neoblasts are marked
TEXTURES OF NATURALLY DEFORMED METAGABBROS 9 9 b, o
1,0 0.9
Core of porphyroctast Rim of porphyroclast Core of neoblast Rim of neoblast
I
1.0--
Magnesio-homblende
Tremolite Aclinolite
9
Tschermakite
9 A 9 A,~ Ultramylonite(E3)
foo:o
I ]-schermakite
Mylonite (E2)
I Core of yroclast
o
~
Magnesio-hornblende
Tremolite Actinolile
0.9
109
o
!
~rrotschermakite
5.75
Tschermakite
i Protomylonite (El) Core of
Ai
porphyroclast
o irrotschermakite } . - - - -
[
__Mg~ (Mg2++Fe2+)
{
....
5,75
! i I
O.C Ferro-actinolite 7.50 8.00
6.50
1.0 i 0.9i
1.0 I 0.9
5,75
Magnesio-hornblende
Tremolite Actinolite
Magnesio-hornblende
Tremolite Actino}ite
~,~.~ o o ,.1~,~
~ . . _ ~
oO~oo ....
_
~ o o 0,9o--
Ferrotschermakite -
(Mg2++Fe2+)[
8.00
Tschermakite ~
Tschermakite
Porphyroclast 2
:00
oo
Banded mylonite (W2) Porphyroctast 1
_Ng~
Ferrotschermakite i
Ferro-hombtende
~.50
5,75
Augen mylonite (Wl) Ferro*actinolite
Ferro-homblende
7.50)
Ferrotschermakite
6.50
5.75
4
Fig. 10. Composition of amphiboles in (a) the eastern and (b) the western belts. (Classification of Leake et was used, (Na + K)A < 0.5.)
by sharp compositional differences expressed by decreasing Si and XMg contents. In the banded mylonite (W2), larger grains show higher Si content than small neoblasts (very low Si content), both being of tschermakitic composition. Temperature (T) was calculated using the hornblende- plagioclase thermometer by Holland & Blundy (1994; thermometer B). Estimated temperatures are based on 30 and 40 amphibole-plagioclase couples in the eastern
al.
(1997)
and western belt, respectively. In the eastern (lower) sheet, calculated T was estimated to be 650 4-50~ while T in the west was 750 + 50 ~ Deformation is interpreted to have taken place under or close to these metamorphic conditions. Pressure (P) could not be estimated as garnet is absent in the mineral assemblage of the metagabbros. However, P of 9 kbar was estimated at the lower contact between the metagabbro sheet and tonalitic sill, where garnets have been formed (St/pskfi e t al. 2001).
110
L. BARATOUX ETAL.
r~
t'-,I
MdMdddd~dd r~
< r.-~
t"-
m
9
~ d ~ d d d d d d ~
TEXTURES OF NATURALLY DEFORMED METAGABBROS
111
e4
M
o ('4
,,.0 ~
I'~ ~0 oo e4 oo ~1- P--. e~h
~"~ ',4~ ~
',~ t"q ~0 ',,0 0", tt3 c4 0", ,--~
9
~-~.~
~8
4-
~dd
dd,~,~d~
~dd,~ddd4~ddd II
-F +
d 4-
o .1...o
0
~
~8 r~
.~
. ~
.qq
~
. q ~
. ~
.~
L. BARATOUX ET AL.
112
45 ixm (El) (Fig. l la). Matrix grains in mylo-
Quantitative textural analysis
Grain size distribution The average grain sizes are presented as median values of the Ferret diameter and the grain size spread is expressed by the difference of third and first quartile (Q3- Ql). The statistical values of the quantitative microstructural and grain size analyses are summarized in Table 4.
Plagioclase.
Plagioclase in XZ section of the
protomylonitic stage in the eastern metagabbro belt is characterized by an average grain size of
nites are marked by an increase of the average grain size, reaching 74 pore in less (E2-pll) and 58 Ixm in more (E2-p12) strained domains. Grain coarsening in the mylonite is consistent with the syndeformational growth of grains originated from subgrain rotation in the protomylonitic stage. This is supported by a greater amount of larger grains in grain size frequency histograms. The grain size is r e d u c e d to ~ 4 5 Ixm in the ultramylonite (E3). Plagioclase grains in the augen mylonite of the western metagabbro belt (W1) are 43 Ixm on average, which is the statistical value
Table 4, Statistical values of the quantitative textural analysis Eastern belt
Western belt
Protomyl. Mylonite Mylonite Ultramyl. Augen myl. Banded myl. Banded myl. pl amp, pll p12 amp, pl amp, pl amp, pll p12 GBPO Eigenvalue r = el/e2 Eigenvector V1 Orientation (~
Pl-pl 1.23 Amp-amp Amp-plt 1.41 Pl-pl 7 Amp-amp Amp-pl * 16 SPO P1 1.45 Eigenvalue Amp r = el/ea Amp t 1.81 Eigenvector V1 P1 - 14 Orientation (~ Amp Amp t 18 Aspect ratio P1 1.59 (median) Amp Amp ~ 2.67 Grain size-Ferret diameter Median (Ixm) P1 Amp Amp t Q1 (~m) P1 Amp Amp t Q3 (~m) P1 Amp Amp t Q3 - QJ (~m) P1 Amp Amp t Skewness "+ PI Amp Kurtosis * P1 Amp R2 P1 correlation Amp coef.
45
1.30 2.50 1.94 24 11 12 1.46 1.83 2.67 21 16 9 1.56 4.46 2.59
1.58 2.60 23 18 1.73 3.43 22 21 1.82 3.02
1.44 3.53 2.18 17 2 7 1.85 4.40 3.26 8 2 8 1.84 3.79 3.00
74 58 45 49 41 49 39 43 33 33 52 43 30 36 30 38 29 31 24 63 103 80 72 68 62 64 52 60 44 30 51 37 42 33 32 26 23 29 20 1.44 - 0.34 - 0.08 - 0.09 0.62 0.25 7.30 2.86 2.77 2.53 4.86 3.12 0.9151 0.9909 0.9979 0.9951 0.9768 0.9942
1.39 2.26 1.55 3 4 -5 1.86 2.89 1.96 1 8 - 1 1.68 2.36 2.28
1.24 1.96 1.50 - 3 - 2 1 1.38 2.21 1.76 2 - 2 9 1.50 2.12 1.84
43 119 59 30 62 34 63 211 102 32 150 68 1.15 0.02 6.99 2.43 0.9442 0.9899
92 94 50 57 61 35 147 142 74 90 81 39 0.01 0.36 2.98 2.74 0.9977 0.9883
*Positive values are oriented anticlockwisewith respect to the horizontal line. tin plagioclasedomains. *The R2 values, skewnessclose to zero, and kurtosis close to 3 signify well-fittedlognormaldistribution.
1.59 5 1.64 8 1.72
72 50 106 56
TEXTURES OF NATURALLY DEFORMED METAGABBROS
113
b) Amphibole ~.. ~ ~'-----2~--~--
'
~
'Plagioclase ~ East '
I.
Amphibole, Amphibolein plagioclasedomains
ttl
o
\
m
Protomylonite - E1
Amphibole - East
~
Mylonit~-E2
[~
Ultramylonite-E3
b
3
~I
~
Mylonite-E2
91 1 ~
I
I Ultramylonite-E3
A Plagioclase
--
5
0
0
50 100 150 Grain size - Ferret diameter (pm)
9 "~ A LU z~ /x
Protomylonite- E1 Mylonite- E2-pll Mylonite- E2-pl2 Ultramylonite- E3
~9 r
"
t
2E
Amphiboleinplagioclase domains
9 Augen mylonite- Wl o Banded mylonite- W2-pll o Banded mylonite- W2-p12
200
18 16 14
fillL. .
Plagioclase
llilJL -grained p l a g i o c l a s e
~8 6 4 2 0
Amphibole - West
- West
~21]il
-
dlllllllllllllllilLlUL J~nn200n~l][]nl250IJ]n~3000 50 100 150 Grain size - Ferret diameter (~m)
50
m
AugenmyloniteW1
[~
Bandedmylonite- W2
100 150 200 250 300 350 400
Grain size - Ferret diameter (p.m)
Fig. 11. Grain size evolution in the eastern (triangles) and western (circles) metagabbros. Degree of defomlation increases from dark to white colour. (a) Calculatedmedians and standard deviations. (b) Histograms showing the characteristic frequenciesof grain size for each deformational stage.
representing the size of the recrystallized matrix grains. Plagioclase grain size is substantially higher in the coarse-grained domain (W2-pll) of the banded mylonite (92 ~zm). The grain size is reduced to 72 ~m on average in the highstrain domain (W2-p12). The grain size distributions (Fig. 1 lb) in protomylonites of the eastern belt (El) show slightly bimodal distribution due to presence of large porphyroclasts surrounded by small recrystallized grains. A higher amount of large grains characterizes the western banded mylonite (W2) compared to the augen mylonite (W1) and eastern mylonites (El-E3). Lognormal distributions correlate with intensity of deformation in both belts.
Amphibole. The amphibole grain size is 49 p~m in the eastern metagabbro mylonite (E2) and 41 ~m in the ultramylonite (E3). However, the grain size spread is always lower than that of plagioclase in the same rock. The size of amphibole grains included within plagioclase domains decreases systematically with increasing deformation, from 49 ~m in the least deformed sample (El) to 39 ~m and 43 ~m in the mylonite (E2-pll and E2-p12, respectively) and 33 p~m in the ultramylonite (E3). The grain sizes are always slightly lower than those of amphibole matrix grains in the same thin section.
Amphibole grain size in the western metagabbro belt is highest in the sample including porphyroclasts (W1), reaching 119 Fm on average. Very high variance is typical of this mylonite, which is due to a relatively low number of new grains and a high number of large porphyroclasts. In the banded mylonites (W2), grain size decreases to 94 Fm on average. In the eastern belt, amphibole grain size distributions (Fig. 11b) are similar to those of plagioclase. They are characterized by well-developed lognormal distribution in both mylonite and ultramylonite stages. In contrast, the distribution of amphibole grain size in the western belt exhibits weaker fits of lognormal distribution with increasing deformation, which is attributed to a higher number of large grains of variable size (Table 4).
Shape preferred orientation PIagioclase. Recrystallized plagioclase grains in the protomylonite of the eastern belt (El) and mylonite (E2-pll) have aspect ratio (AR) of 1.59 and 1.56, respectively (Fig. 12a). Both aspect ratio (AR = 1.82) and SPO increase in the analysed high-strain plagioclase aggregate adjacent to the rigid amphibole porphyroclast in mylonite (E2-pl2). The ultramylonite (E3) exhibits further strengthening of the SPO and similar high aspect ratios (AR = 1.84).
L. BARATOUX ETAL.
114
Plagioclase grains in augen mylonites of the western belt (W l) have the strongest SPO but similar aspect ratios (AR = 1.68). The SPO decreases significantly in the coarse-grained plagioclase layers of banded mylonite (W2-pll), which also have the lowest aspect ratio of all analysed samples (AR = 1.50). Both aspect ratio and SPO further increase in the high-strain domain of banded mylonite (W2-p12).
4 1 plagioclase
3 o ._
I
F] West
o &
'
,
[dRc/dzl
(4)
which places an upper bound on the average shear strength along the perimeter of the crustal slice. For D = 3 0 k m and 6 = 3 0 ~ these bounds are approximately 30 and 20 MPa for low-density and high-density crust, respectively. Additionally, the condition that the crossover of buoyancy and resistive forces has to be reached at a depth of about 100 km further reduces these upper limits (between 20 and 13 MPa, respectively; see Fig. 10). These limits decrease by a factor equal to D/Hc if the detached slice is considerably thinner than the crust, as is usually the case (typically, 5 - 1 0 km thick; Ernst 2001). However, in the above argument it has been assumed that detachment occurs simultaneously along the whole downdip length of the slice, which is certainly an overestimate. A reduction by a factor f - - D / H c of the downdip length of the detached slice reduces the total frictional resistance by the same amount as the positive buoyancy. Consequently, the upper bound of 10-20 MPa on the shear stress is a reasonable estimate. It is reasonable to assume that the rheology of the exhumed material is controlled by the volumetrically predominant continental rocks. Both plagioclase- and quartz-rich rocks are soft at high temperature, with creep strengths < 1 0 M P a at T c. 850~ for strain rates _1 cm a-1 and dip angle > 20 ~ requires the subduction of continental material to depths > 100 km before the change in buoyancy becomes sufficient to overcome its strength. Break-off can occur at shallower depths (>35kin) if the continental lithosphere arriving at the trench is relatively warm (van de Zedde & Wortel 2001). However, the depth of continental subduction in the Ulten case shows that break-off occurred, if at all, in the oceanic part of the slab at z > 100 kin. Models of slab break-off show that the consequent uplift affects a wide region, with maximum
CONTINENTAL SUBDUCTION AND EXHUMATION surface uplift of the order of a few km (Buiter et al. 2002), and that uplift rates (comparable with exhumation rates if erosion keeps approximate pace with uplift) are low ( < l mm a -1 as an order of magnitude; Giunchi et al. 1996). If exhuming crustal slices had reached approximate gravitational equilibrium in the midcrust before break-off, the subsequent exhumation rate would therefore be one order of magnitude lower than before, as observed in the Ulten m61ange. The slow cooling rate during this stage may be attributed to the depth difference between the slices and the level of the break-off. For a material with thermal diffusivity K c. 10 -6 m 2 s-1, such as the lithosphere, the characteristic diffusion time of a thermal perturbation over a distance of several tens of km is of the order of 100 Ma. Using a simple one-dimensional model, Davies and von Blanckenburg (1995) estimate that the temperature increase over 50 Ma at depth z = 20 km would be of the order of 100 ~ if temperature is raised instantaneously to asthenospheric values at z = 80 km. Therefore, both uplift rate and cooling rate of the Ulten m61ange during slow exhumation are compatible with a relatively deep slab break-off. Evidence for the late timing of slab break-off may be related to the production of trondhjemitic melts, which have formed dykes and veins that cut the partially migmatized continental basement. The geochemistry of these trondhjemites excludes melting in situ, and exhibits mixed lower crustal and mantle affinity (Martin et al. 1998). Their age of formation is uncertain, but available results point to c. 300 Ma or soon afterwards (Del Moro et al. 1999), which could date slab break-off and consequent melting of rocks that came in contact with asthenospheric material. Removal of a cool lithospheric root, either by break-off or by convective instability, results in deviatoric tensional stresses in the overlying lithosphere (e.g. Fleitout & Froidevaux 1982). Consequently, slab break-off may have generated a sequence of events involving not only cessation of subduction and slowing down of exhumation, but also lithospheric extension and the widespread Permo-Triassic magmatism of the Eastern Alps.
Concluding remarks In our view, the Ulten m61ange formed by entrainment of mantle wedge peridotites in subducted felsic continental basement. The emplacement of the peridotites occurred at, or shortly before, the attainment of peak conditions. Continental material, therefore, had been subducted to
171
a depth z c. 100 km by 330 Ma. For subduction dip angles and velocities in the ranges 2 0 - 4 0 ~ and 1-5 cm a -~, respectively, this implies that continental subduction had gone on for 1030 Ma before formation of the mrlange. A kinematic model of oceanic and continental subduction proves that this is feasible, provided that the total downdip length of the slab (including the oceanic part) is approximately a factor of two or more larger (depending mainly on the density and thickness ratios) than the downdip length of subducted continental crust. After 330 Ma, the Ulten m~lange underwent a common pre-Alpine exhumation history. For the first 30 Ma, exhumation was fast (c. 0.5 cm a-a along the subduction channel). Similar, or even faster, exhumation rates have been inferred elsewhere, for instance in the Dora Maira Massif, Western Alps (Michard et al. 1993; Rubatto & Hermann 2001), the Himalayas (Chemenda et al. 2000; Treolar et al. 2003), and the Sulu region of China (Ji et al. 2003). The evidence can be accounted for by assuming upward movement of crustal slices along the subduction channel, driven by the upward buoyancy of crustal material. A rough force balance calculation confirms that this evolution is compatible with the theological properties of the materials involved. There is no need to invoke slab break-off at this stage, and there is no evidence that subduction had stopped. The fast exhumation stage brought the Ulten m~lange to depths of c. 25 km. Exhumation and cooling rates decreased by one order of magnitude during the final stages of its pre-Alpine history, coinciding with Permo-Triassic lithospheric extension, rifting, and magmatism. Such two-stage exhumation histories are common in Variscan and Alpine orogens (e.g. Spalla et al. 1996; Duch~ne et al. 1997; Lardeaux et al. 2001). Slab break-off during this stage is compatible with the evidence, and requires a relatively deep level of detachment. In conclusion, we think that the pre-Alpine evolution of the Ulten rocks provides a clear record of subduction of continental crustal material to depths of at least 100 kin. It also suggests strongly that fast exhumation preceded slab detachment, and occurred by buoyancydriven extrusion (as proposed by Chemenda et al. 1995, 1996; Ernst et al. 1997; Ernst 2001). The following slow stage was related to slab break-off (as documented by the occurrence of trondhjemitic melts; Del Moro et al. 1999) and consequent isostatic uplift and lithospheric extension. This sequence of events is compatible both with the evidence and with the rheological properties of the materials.
172
G. RANALLI ETAL.
The research of the senior author is financed by a discovery grant from NSERC (Natural Sciences and Engineering Research Council of Canada), which has also provided support for R. Mahatsente's Postdoctoral Fellowship. We thank G. V. Dal Piaz, G. Godard and P. Nimis for many discussions on Alpine geology over the years. Thorough reviews by T. V. Gerya and F. Gueydan helped with the revisions.
References BOUSQUET, R., GOFFE, B., HENRY, P., LE PICHON, X. & CHOPIN, C. 1997. Kinematic, thermal and petrological model of the Central Alps: Lepontine metamorphism in the upper crust and eclogitisation of the lower crust. Tectonophysics, 273, 105-127. BREY, G. P. & KOHLER, T. 1990. Geothermobarometry in four-phase lherzolites, II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology, 31, 1352-1378, BRUEC~:NER, H. K. 1998. Sinking intrusion model for the emplacement of garnet-bearing peridotites into continent collision orogens. Geology, 26, 631-634. BUITER, S. J. H., GOVERS, R. & WORTEL, M. J. R. 2002. Two-dimensional simulations of surface deformation caused by slab detachment. Tectonophysics, 354, 195-210. BUROV, E., JOLIVET, L., LE POURHIET, L. & POLIAKOV, A. 2001. A thermomechanical model of exhumation of high pressure (HP) and ultrahigh pressure (UHP) metamorphic rocks in Alpine-type collision belts. Tectonophysics, 342, 113-136. CHEMENDA, A. I., MATTAUER,M., MALAVIEILLE,J. & BOKt;N, A. N. 1995. A mechanism for syncollisional rock exhumation and associated normal faulting: results from physical modelling. Earth and Planetary Science Letters, 132, 225 -232. CHEMENDA, A. I., MATTAUER, M. & BOKUN, A. N. 1996. Continental subduction and a mechanism for the exhumation of high-pressure metamorphic rocks: new modelling and field data from Oman. Earth and Planetary Science Letters, 143, 173-182. CHEMENDA, A. I., BURG, J.-P. & MATTAUER, M. 2000. Evolutionary model of the Himalaya-Tibet system: geopoem based on new modeling, geological and geophysical data. Earth and Planetary Science Letters, 174, 397-409. CHOPIN, C. 1984. Coesite and pure pyrope in highgrade blueschists of the Western Alps: a first record and some consequences. Contributions to Mineralogy and Petrology, 86, 107-118. CHOPIN, C. 2003. Ultrahigh pressure metamorphism: tracing continental crust into the mantle. Earth and Planetary Science Letters, 212, 1-14. CLOOS, M. & SHREVE,R. L. 1988. Subduction-channel model of prism accretion, melange formation, sediment subduction, and subduction erosion at convergent plate margins. Pure Applied Geophysics, 128, 455-499.
DAL PIAZ, G. V. 1993. Evolution of Austroalpine and Upper Penninic basement in the northwestern Alps from Variscan convergence to post-Variscan extension. In: VON RAUMER,J. F. & NEUBAUER,F. (eds) The Pre-Mesozoic Geology in the Alps. Springer, Berlin, 327-343. DAL PIAZ, G. V. 1999. The Austroalpine-Piedmont nappe stack and the puzzle of Alpine Tethys. Memorie Scienze Geologiche, Universit?t di Padova, 51, 155-176. DAVIES, J. H. & v o n BLANCKENBURG,F. 1995. Slab breakoff: a model of lithosphere detachment and its test in the magmatism and deformation of collisional orogens. Earth and Planetary Science Letters, 129, 85-102. DEL MORO, A., MARTIN, S. & PROSSER, G. 1999. Migmatites of the Ulten zone (NE Italy), a record of melt transfer in the deep crust. Journal of Petrology, 40, 1803-1826. DRURY, M. R. & FITZ GERALD, J. D. 1998. Mantle rheology: insights from laboratory studies of deformation and phase transitions. In: JACKSON, I. (ed) The Earth's Mantle - Composition, Structure, and Evolution. University Press, Cambridge, 503 -559. DUCHI~NE, S., LARDEAUX, J. M. & ALBARI~DE, F. 1997. Exhumation of eclogites: insights from depth-time path analysis. Tectonophysics, 280, 125-140. ERNST, W. G. 2001. Subduction, ultrahigh-pressure metamorphism, and regurgitation of buoyant crustal slices - implications for arcs and continental growth. Physics of Earth and Planetary Interiors, 127, 253-275. ERNST, W. G. & PEACOCK, S. M. 1996. A thermotectonic model for preservation of ultrahigh-pressure phases in metamorphosed continental crust. In: BEBOUT, G. E., SCHOLL, D. W., KIRBY, S. H. & PLATT, J. P. (eds) Subduction - Top to Bottom. American Geophysical Union, Geophysical Monograph, 96, 171-178. ERNST, W. G., LtOU, J. G. & COLEMAN, R. G. 1995. Comparative petrotectonic study of five Eurasian ultrahigh-pressure metamorphic complexes. International Geology Review, 37, 191 - 21 I. ERNST, W. G., MARUYAMA, S. & WALLIS, S. R. 1997. Buoyancy-driven, rapid exhumation of ultrahighpressure metamorphosed continental crust. Proceedings of the U. S. National Academy of Sciences, 94, 9532-9537. FLEITOUT, L. & FROIDEVAUX,C. 1982. Tectonics and topography for a lithosphere containing density heterogeneities. Tectonics, 1, 21-56. GERYA, T. V., STt3CKHERT, B. & PERCHUK, A. L. 2002. Exhumation of high-pressure metamorphic rocks in a subduction channel: a numerical simulation. Tectonics, 21, 1056, doi:10.1029/ 2002TC001406. GIUNCHI, C., SABADINI, R., BOSCHI, E. & GASPERINI, P.,1996. Dynamic models of subduction: geophysical and geological evidence in the Tyrrhenian Sea. Geophysical Journal International, 126, 555-578.
CONTINENTAL SUBDUCTION AND EXHUMATION GODARD, G. 1983. Dispersion tectonique des 6clogites de Vend6e lors d'une collision continentcontinent. Bulletin de Mindralogie, 106, 719-722. GODARD, G. & MARTIN, S. 2000. Petrogenesis of kelyphites in garnet peridotites: a case study from the Ulten zone, Italian Alps. Journal of Geodynamics, 30, 117-145. GODARD, G., MARTIN, S., PROSSER, G., KIENAST,J. R. & MORTEN, L. 1996. Variscan migmatites, eclogites and garnet peridotites of the Ulten zone, Eastern Austroalpine system. Tectonophysics, 259, 313-341. HARLEY, S. L. 1984. An experimental study of the partitioning of Fe and Mg between garnet and orthopyroxene. Contributions to Mineralogy and Petrology, 86, 359-373. HIBBITT, KARLSSON& SORENSENInc. 2001. ABAQUS User's Manual: Version 6.3. HKS, Pawtucket, Rhode Island. HONDA, S. 1985. Thermal structure beneath Tohoku, northeast Japan - a case study for understanding the detailed thermal structure of the subduction zone. Tectonophysics, 112, 69-102. HOOGERDUIJN STRATING, E. H. & VISSERS, R. L. M. 1991. Dehydration-induced fracturing of eclogitefacies peridotites: implications for the mechanical behaviour of subducting oceanic lithosphere. Tectonophysics, 200, 187-198. HYNDMAN, R. D. & WANG, K. 1993. Thermal constraints on the zone of major thrust earthquake failure: the Cascadia subduction zone. Journal of Geophysical Research, 98, 2039-2060. JI, S., SARUWATARI, K., MA1NPRICE, D., WIRTH, R., XU, Z. & X1A, B. 2003. Microstructure, petrofabrics and seismic properties of ultra high-pressure eclogites from Sulu region, China: implications for rheology of subducted continental crust and origin of mantle reflections. Tectonophysics, 370, 49-76. LARDEAUX, J. M., LEDRU, P., DANIEL, I. & DUCn~NE, S. 2001. The Variscan French Massif Central - a new addition to the ultra-high pressure metamorphic club: exhumation processes and geodynamic consequences. Tectonophysics, 332, 143-167. MAHATSENTE, R. & RANALLI, G. 2004. Time evolution of negative buoyancy of an oceanic slab subducting with varying velocity. Journal of Geodynamics, 38, 117-129. MANCKTELOW, N. S. 1995. Nonlithostatic pressure during sediment subduction and the development and exhumation of high pressure metamorphic rocks. Journal of Geophysical Research, 100, 571-583. MARTIN, S. 2003. Tectonic setting and pre-Alpine evolution of the Tonale Nappe, Eastern Austroalpine. Memorie Scienze Geologiche, Universitit di Padova, 54, 167-170. MARTIN, S., GODARD, G., PROSSER, G., SCHIAVO, A., BERNOULLI, D. & RANALLI, G. 1998. Evolution of the deep crust at the junction Austroalpine/ Southalpine: the Tonale nappe. Memorie Scienze Geologiche, Universitb di Padova, 50, 3-50.
173
MICHARD, A., CHOPIN, C. & HENRY, C. 1993. Compression versus extension in the exhumation of the Dora-Maira coesite-bearing unit, Western Alps, Italy. Tectonophysics, 221, 173-193. MILLER, C., SATIR, M. & FRANK, W. 1980. Highpressure metamorphism in the Tauern window. Mitteilungen der Osterreichischen Geologischen Gesellschaft, 71/72, 89-97. NEUBAUER, F. & VON RAUMER,J. F. 1993. The Alpine basement - linkage between the Variscides and East-Mediterranean mountain belts. In: YON RAUMER, J. F. & NEUBAUER, F. (eds) The PreMesozoic Geology in the Alps. Springer, Berlin, 641-661. NIMIS, P. & MORTEN, L. 2000. P-T evolution of 'crustal' garnet peridotites and included pyroxenites from the Nonsberg area (Upper Autroalpine), NE Italy: from the wedge to the slab. Journal of Geodynamics, 30, 93-115. NTAFLOS, T., THONI, M. & YIN, Q. 1993. Geochemie der Ultentaler ultramafitite. Mitteilungen der Osterreichischen Geologischen Gesellschaft, 138, 169-178. OBATA, M. & MORTEN, L. 1987. Transformation of spinel lherzolite to garnet lherzolite in ultramafic lenses of the Austridic crystalline complex, northern Italy. Journal of Petrology, 28, 599-623. OLESKEVICH, D. A., HYNDMAN, R. D. & WANG, K. 1999. The updip and downdip limits to great subduction earthquakes: thermal and structural models of Cascadia, south Alaska, SW Japan, and Chile. Journal of Geophysical Research, 104, 14965-14991. O'NEILL, H. S. C. & WOOD, B. J. 1979. An experimental study of F e - M g partitioning between garnet and olivine and its calibration as a geothermometer. Contributions to Mineralogy and Petrology, 70, 59-70. PEACOCK, S. M. 1996. Thermal and petrologic structure of subduction zones. In: BEBOUT, G. E., SCHOLL, D. W., KIRBY, S. H. & PLATT, J. P. (eds) Subduction - Top to Bottom. American Geophysical Union, Geophysical Monograph, 96, 119-133. RANALLI, G. 1995. Rheology of the Earth, 2nd edn. Chapman & Hall, London. RANALLI, G. 2000. Rheology and subduction of continental lithosphere. In: RANALLI,G., RICCI, C. A. & TROMMSDORF, V. (eds) Crust-Mantle Interactions. Proceedings Siena International School of Earth and Planetary Science, 21-40. RANALLI, G. 2003. A model of Palaeozoic subduction and exhumation of continental crust: the Ulten Unit, Tonale Nappe, Eastern Austroalpine. Memorie Scienze Geologiche, Universith di Padova, 54, 205-208. RANALLI, G., PELLEGR1NI, R. & D'OFFIZI, S. 2000. Time dependence of negative buoyancy and the subduction of continental lithosphere. Journal of Geodynamics, 30, 539-555. REGARD, V., FACCENNA, C., MARTINOD, J., BELLIER, O. & THOMAS, J.-C. 2003. From subduction to collision: control of deep processes on the evolution of convergent plate boundary.
174
G. RANALLI ET AL.
Journal of Geophysical Research, 108, 2208, doi: 10.1029/2002JB001943. RING, U., BRANDON, M. T., WILLETT, S. D. & LISTER, G. S. 1999. Exhumation processes, ln: RING, U., BRANDON, M. T., LISTER, G. S. & WILLETT, S. D. (eds) Exhumation Processes: Normal Faulting, Ductile Flow and Erosion. Geological Society, London, Special Publications, 154, 1-27. RUBATTO, D. & HERMANN, J. 2001. Exhumation as fast as subduction? Geology, 29, 3-6. SHREVE, R. L. & CLOOS, M. 1986. Dynamics of sediment subduction, melange formation, and prism accretion. Journal of Geophysical Research, 91, 10229-10245. SPALLA, M. I., LARDEAUX, J. M., DAL PIAZ, G. V., Gosso, G. & MESSIGA, B. 1996. Tectonic significance of Alpine eclogites. Journal of Geodynamics, 21, 257-285. THOMPSON, A. B., SCHULMANN,K. & JEZEK, J. 1997. Thermal evolution and exhumation in obliquely convergent (transpressive) orogens. Tectonophysics, 280, 171-184. THONI, M. 1999. A review of geochronological data from the Eastern Alps. Scweizerischen Mineralogie und Petrographie Mitteilungen, 79, 209-230. TREOLAR, P. J., O'BRIEN, P. J., PARRISH, R. R. & KHAN, M. A. 2003. Exhumation of early Tertiary, coesite-bearing eclogites from the Pakistan Himalaya. Journal of the Geological Society, London, 160, 367-376. TUMIATI, S. 2002. Le peridotiti a granato di Cima Vedetta Alta (Val d'Ultimo, Alto Adige). Thesis, University of Padova.
TUMIATI, S., THONI, M., NIMIS, P. & MARTIN, S. 2003. Mantle-crust interactions during Variscan subduction in the Eastern Alps (Nonsberg-Ulten zone): geochronology and new petrological constraints. Earth and Planetary Science Letters, 210, 509-526. VAN DEN BEUKEL, J. 1992. Some thermomechanical aspects of the subduction of continental lithosphere. Tectonics, 11, 316-329. VAN DEN BEUKEL, J. & WORTEL, M. J. R. 1988. Thermomechanical modeling of arc-trench regions. Tectonophysics, 154, 177-193. VAN DE ZEDDE, D. M. A. & WORTEL, M. J. R. 2001. Shallow slab detachment as a transient source of heat at midlithospheric depths. Tectonics, 20, 868-882. YON RAUMER, J. F. & NEUBAUER, F. 1993. Late Precambrian and Palaeozoic evolution of the Alpine basement - an overview. In: YON RAUMER, J. F. & NEUBAUER, F. (eds) The PreMesozoic Geology in the Alps. Springer, Berlin, 625 -639. WANG, Z. & Jt, S. 1999. Deformation of silicate garnets: brittle-ductile transition and its geological implications. Canadian Mineralogist, 37, 525-541. WANG, Z. & JI, S. 2000. Diffusion creep of finegrained garnetite: implications for the flow strength of subducting slabs. Geophysical Research Letters, 27, 2333-2336. WYLLIE, P. J. 1988. Magma genesis, plate tectonics, and chemical differentiation of the Earth. Reviews of Geophysics, 26, 370-404.
Kinematics of syneclogite deformation in the Bergen Arcs, Norway: implications for exhumation mechanisms H U G U E S R A I M B O U R G 1, L A U R E N T J O L I V E T l, L O I C L A B R O U S S E l, YVES LEROY 2 & DOV AVIGAD 3
1Laboratoire de Tectonique, UMR 7072, Universitd Pierre et Marie Curie, T 2 6 - 0 El, case 129, 4 place Jussieu, 75252 Paris cedex 05, France (e-mail: hugues, raimbourg @ lgs.jussieu.fr) 2Laboratoire de Mdcanique des Solides, Ecole Polytechnique, Palaiseau, France 3Institute of Earth Sciences, Hebrew University of Jerusalem, Jerusalem, Israel Abstract: The northwestern part of Holsncy island, in the Bergen Arcs, Norway, consists of
a granulite-facies protolith partially transformed at depth in eclogite (700 ~ > 19 kbars) and amphibolite (650 ~ 8-10 kbars) facies during the Caledonian orogenesis. Eclogitized zones are mainly planar objects (fractures with parallel reaction bands and cm-to100 m-scale shear zones). Eclogitic zones are distributed in two sets of orientations and the associated deformation can be described as 'bookshelf tectonics'. The major shear zones strike around N120 and dip to the North, and show consistent top-to-the-NE shear sense throughout the area. In the large-scale kinematic frame of Caledonian NW-dipping slab, eclogitic shear zones are interpreted as the way to detach crustal units from the subducting slab and to prevent their further sinking. As the retrograde amphibolitic deformation pattern is similar to the eclogitic one, the detached crustal units started their way up along these eclogitic shear zones. Radiometric ages of eclogitic and amphibolitic metamorphism and their comparison with the chronology of Caledonian orogenesis show that the deformation recorded on HolsnCy occurred in a convergent context. The mechanism we propose can thus account for the first steps of exhumation during collision.
Relics of high pressure (HP) and ultra high pressure (UHP) parageneses record the passage at depth of crustal units (Chopin 1984; Smith 1984; Wain 1997; Chopin & Schertl 2000; Ernst & Liou 2000; Jolivet et al. 2003). Such witnesses of a deep history, found in most current or past collision zones, can provide constraints on the deep structure and time evolution of the collision. In the Norwegian Caledonides, as well as in the Alps or in the Himalayas, plate collision was preceded by the closure of an ocean by subduction. In this context, oceanic and then continental crust can be dragged down to large depths. Downpull by the underlying lithospheric mantle in a subducting slab is indeed a widely accepted mechanism for burial of crust. On the contrary, the processes involved in the exhumation of buried tectonic units are not well understood, and several models have been designed in order to account for the return path to the surface (see review by Platt 1993). Exhumation can result from a drastic change in either vertical or horizontal boundary conditions. Andersen
et al. (1991) propose that the exhumation of HP units in the Caledonides is achieved through the eduction of an orogenic wedge, caused by the detachment of the lithospheric mantle root followed by a large-scale isostatic rebound. Alternatively, deeply buried rocks can be brought back to the surface in a reverse movement, as a result of plate kinematics change from convergence to divergence (Fossen 1992). A comparison of the ages of HolsnCy eclogites and amphibolites given by Boundy et al. (1992) and Ktihn (2002), with the ages of the eclogites in the nearby Western Gneiss Region (WGR) (review by Torsvik & Eide 1998; Carswell et al. 2003; Tucker et al. in press), indicates that Holsncy eclogites were exhuming at the same time as WGR was being buried ( 4 1 0 400 Ma). The first steps of HolsnCy eclogites exhumation occurred while the collision was still active, therefore it can only be accounted for by collisional models. Most collisional models, such as corner or wedge flow models (Jischke 1975; England &
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 175-192. 0305-8719/05/$15.00
(c) The Geological Society of London 2005.
176
H. RAIMBOURG ET AL.
Holland 1979; Cloos 1982; Shreve & Cloos 1986; see review by Mancktelow 1995), as well as analogue models (Chemenda et al. 1995), do not take into account rheological changes due to metamorphic reactions. The two-dimensional model of Burov et al. (2001) does consider metamorphic reactions caused by rock burial, but the reactions depend only on the P - T conditions and are instantaneous; that is, 100% of crustal material is transformed to eclogite as soon as the boundary of the P - T eclogitic field is crossed. The simple one-dimensional isostatic model of Dewey et al. (1993) takes into account firstorder reaction kinetics. As demonstrated by Austrheim (1987), eclogitization reactions need some fluid input to occur. As a result, the fluid supply determines which fraction of the granulitic unit can be transformed into eclogite, and the fluid transport limits the transformation rate. Eclogitic metamorphic reactions depend not only on variations in P - T conditions but also on fluid availability. Albeit not well accounted for in many models, eclogite-facies reactions are likely to have important consequences on the history of rocks affected: the complete eclogitization of the HolsnCy granulite leads to a density increase of 7%, as well as a viscosity decrease visible in the field (Austrheim 1987). The aim of this study is to analyse the deformation recorded within eclogites on Holsncy and its relations with the tectonic history of the unit during the Caledonian orogenesis. Eclogitefacies deformation is essentially non-coaxial, eclogitic structures showing a 'bookshelf' geometry with a top-to-the-NE sense of shear on the main set of shear zones. We interpret this deformation as the first step towards exhumation. We propose a geometrical model where the eclogitization reactions enable crustal slivers to detach from the downgoing slab along eclogitic shear zones and to start their way up.
Geological setting The Scandinavian Caledonides result from the closure of the Iapetus ocean and the NW-dipping continental subduction of Baltica under Laurentia, in a time spanning from late Ordovician to early Devonian periods. A series of allochtonous nappes, originating from the Iapetus ocean and from the western margin of Baltica were thrust eastward onto the autochtonous shield of Baltica (Krogh 1977; Roberts & Gee 1985). The Bergen Arcs consist of a series of arcuate thrust sheets centred around Bergen, Norway (Fig. 1). The Bergen Arcs nappe pile is in tectonic contact to the east with the parautochtonous WGR through the Bergen Arc shear zone
(BASZ, see Fig. 1). The BASZ involves parts of three tectonic units: WGR, the Kvalsida Gneiss and the Major Bergen Arc zone (Wennberg 1996). Another fault zone, the Main Caledonian Thrust (MCT), constitutes the western border of the Bergen Arcs and separates the structurally lower Oygarden Gneiss Complex from the Minor Arc and Ulriken Gneiss Complex (Fig. 1). The Lind,s nappe lies in the highest tectonic position in the Bergen Arcs nappe pile. This crustal nappe is supposed to be originally a lower crust unit and contains abundant mafic material, from gabbroic anorthosite to pure anorthosite, as well as various rocks of the charnockite suite like mangerite (Kolderup & Kolderup 1940). The northwestern part of HolsnCy island (Fig. 2a), which is the object of our study, belongs to the Lind,s nappe, and is composed of mangerites and anorthositic granulites (Austrheim & Griffin 1985), intensively reworked under eclogite and amphibolite-facies conditions. It is separated from the Meland Nappe (southeastern HolsnCy) by the Rossland Shear Zone (Fig. 2a), which is up to 2.5 km wide and shows a complex metamorphic history from amphibolitefacies to greenschist-facies conditions (Birtel et al. 1998; Schmid et al. 1998). The Meland Nappe consists of granulites with amphibolite to greenschist-facies mylonites, without any evidence of eclogite-facies metamorphism (Birtel et al. 1998).
Field relations on northwestern Holsn0y The geology of the area and its metamorphic history have been studied in detail by Austrheim and co-workers (Austrheim 1987, 1990a, b, 1994; Austrheim & Griffin 1985; Austrheim & MCrk 1988; Austrheim & Engvik 1995; Boundy et al. 1992). During the Caledonian history, the granulitic protolith has recorded successive parageneses on the exhumation path, under eclogite, amphibolite and finally greenschistfacies P - T conditions. Granulite-facies protolith
Northwestern Holsn0y granulitic anorthosites present two different fabrics. In the well-foliated parts, the granulitic foliation is defined by alternating plagioclase-rich and pyroxene-garnetrich layers. In the more coronitic parts, a plagioclase matrix encloses ellipsoidal granulitic coronas, composed of a core of orthopyroxene surrounded by layers of clinopyroxene and garnet. The strong shape fabric of the coronas
KINEMATICS OF ECLOGITE-FACIES DEFORMATION
177
Fig. 1. Geological map of the Bergen Arcs (modified from Ragnhildsveit & Helliksen 1997). General map of Norwegian Caledonides modified from Roberts & Gee (1985). defines a granulitic lineation. Both foliation and lineation were acquired during the 900Ma Grenvillian granulite-facies deformation (Cohen et al. 1988; Bingen et al. 2001). Eclogite-facies metamorphism
The anorthositic complex was partially reworked under eclogite-facies conditions (Fig. 2b). The predominant eclogitic assemblage in anorthosite is omphacite, garnet, kyanite, zoisite, and minor phengite, + rutile, +_quartz, +__amphibole (Boundy et al. 1992). Gabbroic eclogites consist predominantly of omphacite, garnet, and minor phengite, rutile, quartz, _+carbonates. Input of fluid was necessary for eclogitic reactions to occur since both assemblages are hydrous. Therefore a shortage in available fluid resulted in the metastable preservation of part of the
granulite in the eclogitic P-T field. A heterogeneous distribution of the fluid supply enabled different grades of the transformation to be 'frozen in' (Austrheim 1987; Austrheim & MCrk 1988; see Fig. 3). Some parts of the granulitic unit are macroscopically not affected by eclogitization reactions. Elsewhere, the granulite is cut by mm-wide fractures containing hydrous eclogitic minerals and quartz. On both sides of these fractures, a dm-wide dark band corresponds to partially eclogitized granulite. A further progression in the transformation is illustrated by minor eclogitic shear zones. These minor shear zones, 10 cm to 2 m wide and a few tens of metres long, display a well-defined foliation and cut through a coherent skeleton of granulite. Where the density of the shear zones increases, they form an anastomosing network surrounding rounded, 1 to 10 m large blocks of
178
H. RAIMBOURG E T A L .
Fig. 2. (a) Geological map of Holsn~y by Birtel et al. (1998), Boundy et al. (1997a), KiJhn (2002), Ragnhildsveit & Helliksen (1997). Location of Figures 2b, 4 and 9. Coordinates correspond to UTM grid, zone 32:22 ---- 67 22 000, 81 = 2 81 000. (b) Northwestern Holsn~y (Austrheim et al. 1996; Boundy et al. 1997b) with heterogeneous distribution of eclogitic overprint on granulite. Eclogite, Breccia and Granulite correspond to degrees of eclogitic metamorphism >80%, 40-80%, and 14.6 kbars (Boundy et al. 1992), and 700 ~ and 17 kbars (Mattey et al. 1994). Boundy et al. (1996) estimated P - T conditions of 690 ~ and 8-12 kbars for high-grade amphibolite in the northern and eastern areas in the Lind,s Nappe, whereas estimates by Ktihn (2002) on amphibolites on Northwest Holsnr using P - T pseudosections and amphibole thermometry, yield 600 ~ and 8 kbars. The chronology of the H P - L T metamorphism is still debated. Two scenarios can be proposed, corresponding to a metamorphism of Lind,s Nappe early in the Caledonian history, or coeval with the main collisional phase (Scandian phase, dated c. 425 Ma; Torsvik et al. 1996): 9 Ages around 460 Ma for the eclogite metamorphism (U/Pb on sphene and epidote by Boundy et al. 1997a and U/Pb on zircon by Bingen et al. 2001), followed by rapid cooling to 500 ~ at 455-445 Ma (Ar/Ar on hornblende, Boundy et al. 1996), and to 375 ~ at 430 Ma (Ar/Ar on muscovite, Boundy et al. 1996). 9 Ages around 420 Ma for the eclogite metamorphism (419 _+ 4 Ma with U/Pb on zircon on Holsncy, Bingen et al. 1998) consistent with a highly unconstrained age of 421 _+
68 Ma with Rb/Sr mineral isochrone (Cohen et al. 1988). Amphibolite-facies metamor-
phism dating yielded 4 0 9 _ 8 Ma (Rb/Sr isochron age, Austrheim 1990a) and 418 + 9 Ma for the Fonnes trondjhemic dyke intruding Proterozoic rocks under P - T conditions of 8 - 1 0 kbars and 650-700 ~ at Austrheim locality, in the northeast of Lindfis nappe (Kiihn 2002). This short time span between amphibolite and eclogitefacies metamorphism may result either from rapid exhumation or from the fact that, at the same time within Lind,s Nappe, HolsnCy was more deeply buried than Austrheim.
Kinematics of eclogitic deformation Eclogitic lineations
Eclogitic stretching lineations are present in the field at two different scales, as described by Rey et al. (1999) and Boundy et al. (1992). 9 At the mm scale, the orientation of rod-shaped omphacite defines a mineral lineation. 9 At the c m - d m scale, the ellipsoidal granulitic coronas, where eclogitized and deformed, form elongate dark mineral aggregates. At the microscopic scale, crystallographic preferred orientations (CPO) of omphacite show a strong correlation with structural directions (Boundy et al. 1992; Labrousse 2001; Bascou et al. 2002). Strain r e g i m e f r o m p r e v i o u s studies
Boundy et al. (1992) concluded that the strain regime was non-coaxial. This was based on CPO patterns of omphacites from four samples in the Lower Eldsfjellet Shear zone, but the analysis was extremely local and did not yield any sense of shear. A top-to-the-NE sense of shear was inferred from the offset of one metagabbro dyke, from offsets across minor eclogitic shear zones (Boundy et al. 1992) and from asymmetric mineral clusters and crystallization tails (Rey et al. 1999). These observations are nevertheless too few to draw any conclusion at the scale of granulite unit. We therefore extensively collected kinematic indicators in major shear zones, as well as in less transformed areas, to decipher whether consistent kinematics could be found at large scale.
KINEMATICS OF ECLOGITE-FACIES DEFORMATION
Eclogitic deformation in large shear zones Geometry of major shear zones and lineation directions. The granulitic unit is cut by a set of 10 to 100 m large shear zones that can be followed along strike for several hundreds of metres (Fig. 4). These shear zones trend on average between N90 and N120, except east of Skurtveit and in Lower Eldsfjellet, where they swing to strike around N60. All shear zones have a northward dip between 10 and 40 ~ Our collected measurements of lineations and foliations in the major eclogitic shear zones are in agreement
181
with the data of Rey et al. (1999). On average, the lineations in Hundskjeften strike N90, N70 in Upper Eldsfjellet and N40 in Lower Eldsfjellet. The Skurtveit shear zone shows two distinct sets of lineations. The scarcity of data collected on this shear zone prevents us from trying to explain the coexistence of these two sets. In all shear zones, lineations have a northeastward plunge between 10 and 30 ~
Kinematic indicators in major shear zones. Formerly spherical objects, when deformed by
80
85
25
1 km
t
i North 22
::: .:j Eclogite !. . . . . . ; "Breccia" ....................... Granulite tr
} _,,_ 30
Eclogitic Lineation Eclogitic Foliation dip angle
Fig. 4. Eclogitic foliations and lineations in major shear zones. Arrows: sense of shear. Hund., Skurt., L.Elds. and U.Elds. refer to Hundskjeften, Skurtveit, Lower Eldsfjellet and Upper Eldsfjellet, respectively.
182
H. R A I M B O U R G ET AL.
Fig. 5. (a) Sense of shear deduced from asymmetrical shapes of granulite boudins, sigmoidal eclogite foliation around boudins and deflected granulite foliation along boudins borders. Locations on Figure 2b. (b) Sense of shear deduced from S-C structures and from asymmetrically deformed coronas or mafic aggregates. These criteria show consistent top-to-the-NE sense of shear. See locations on Figure 2b.
KINEMATICS OF ECLOGITE-FACIES DEFORMATION
183
Fig. 5. Continued.
simple shear, get an asymmetrical shape (Passchier & Trouw 1998). At the same time the foliation in the matrix wrapping around the objects adopts a sigmoidal morphology. This morphological description, which is scaleindependent, was applied to two different classes of objects on northwestern Holsncy: eclogitized omphacite-rich former coronas embedded in a kyanite-zoisite-rich matrix ( 1 - 1 0 c m in size), and untransformed granulite boudins embedded in eclogite matrix (50 cm to 10 m in size) in eclogitic shear zones (Fig. 5a and b, respectively). The sense of shear can also be inferred from the deformation of the internal granulitic foliation in granulitic boudins. In the relatively narrow transition zone between preserved granulites in the core of the boudin and eclogitic shear zones, the granulitic foliation, where oblique to the eclogitic foliation, is bent to become parallel to the eclogitic foliation, giving shear sense along the border of granulitic boudins (Fig. 5b). The study of large-scale asymmetric features such as granulitic boudins is particularly relevant for kinematics analysis: the progressive formation of eclogitic shear zones tends to isolate large blocks, from one to several tens of metres large, of undeformed and resistant granulite in an eclogitic matrix. The pertinent
characteristic 'grain size' is thus very large and the search for regionally significant kinematic indicators requires observations at a large scale. Both within cm-scale coronas and m-scale boudins, most collected criteria yield top-tothe-NE sense of shear throughout the granulitic unit. This sense of shear is compatible with the overall sigmoidal shape of the eclogite foliation throughout the island. Although asymmetrical objects can be found in every shear zone, their density is highly variable. They are sparse in the Upper and Lower Eldsfjellet shear zones and much more ubiquitous in the Hundskjeften shear zone and to a smaller extent in the Skurtveit shear zone. In order to demonstrate this density we carried out a detailed mapping of a 100 x 100 m area in the Hundskjeften shear zone containing many metric sized boudins (Fig. 6). The asymmetrical shape of most boudins, the sigmoidal eclogitic foliation around them and the deflection of granulitic foliation along boudin borders demonstrate a consistent dextral sense of shear at the map scale. The average eclogitic foliation in this area is N120 40N and the lineation trends N80. The apparent dextral sense of shear results from top-to-the-ENE movement.
184
H. RAIMBOURG ET AL.
Fig. 6. Map of asymmetrical boudins in the Hundskjeften 'breccia' zone showing consistent top-to-the-NE sense of shear. See location on Figure 2b.
Eclogitic deformation in little transformed zones In moderately transformed zones, eclogitization is present only in 1 to 10 cm wide bands, either on both sides of a m m - w i d e fracture or as narrow shear bands without central fracture. A large proportion of eclogitic fractures - that is, ram-wide fractures with 1 to 10 c m wide eclogitic reaction bands on both sides - does not show any brittle nor ductile deformation. These 'static' fractures seem to be randomly oriented. In the rest of the eclogitic bands, the kinematics can be deduced from the offset/ deflection of granulitic foliation/lineation across the eclogitized zone. The orientations of these bands, as well as the deformation that affects them, show a very consistent structural pattern and can be separated in two main sets (Fig. 7a). The first set of bands strikes between N90 and N120, dips northeastward, and displays n o r m a l dextral m o v e m e n t associated with NE-striking lineations. The second set strikes between N30 and N80, dips northwestward, and displays n o r m a l - s i n i s t r a l m o v e m e n t associated with NW-striking lineations. Representative outcrops (Fig. 7b) show either sinistral bands (B), dextral bands (C), or both (A and B'). Where both sets of bands are present, the dextral bands are wider than the sinistral ones. The geometry and kinematics of narrow eclogitic bands suggest a bookshelf mechanism: the rhomboedric blocks of granulite are bounded by a dominant set of normal dextral shear bands and a smaller conjugate set of normal sinistral shear bands (Fig. 8).
(a)
-~--~ ........
N
Shear zones/Fractures with dextral movement Shear zones/Fractures with sinistral movement
Fig. 7. (a) Stereographic plot of the two sets of minor eclogitic shear zones/fractures. The major set (solid lines) corresponds to normal-dextral movement. The minor set (dotted lines) corresponds to normal-sinistral movement. (b) Representative outcrops showing the distribution of eclogitic fractures and narrow shear zones, with two sets of orientations (minor set striking N30-60, with apparent sinistral movement, major set striking N90-120, with apparent dextral movement). Relative movement is deduced either from deflected granulite foliation or from offsets of pyroxene-garnet lenses. Sketch (B) shows shear zones with apparent sinistral movement, sketch (C) shows eclogite-facies shear zones/fractures with apparent dextral movement. Sketches (A) and (B') show the two sets of shear zones/ fractures. See locations on Figure 2b (same location for B and B').
KINEMATICS OF ECLOGITE-FACIES DEFORMATION
185
Fig. 7. Continued.
Moreover, the dextral set of eclogitic bands is comparable in orientation and shear direction with the major shear zones observed in the more eclogitized areas. The bookshelf-type deformation in moderately eclogitized areas is thus
Kinematics of amphibolitic deformation
Fig. 8. Bookshelf geometry with top-to-the-NE major shear zones and conjugate subordinate shear zones related to counterclockwise block rotation. Minor shear zones strike between N30 and N60, with normalsinistral sense of shear. Major shear zones strike between N90 and N120, with normal-dextral sense of shear.
Within the granulitic unit, amphibolite-facies retrogression is less conspicious in the field than the eclogite-facies metamorphism. Amphibolitic shear zones are typically 10 cm to 1 m wide and cannot be followed along strike for more than a few metres. These shear zones, which cross-cut the granulite foliation, strike in average N W - S E with a large scatter (Fig. 9). The deflection of the granulitic foliation near the shear zones gives an average top-to-the-SE sense of shear. The Rossland amphibolitic shear zone, which separates the Meland Nappe from the granulitic unit, is at least a few hundred metres wide and can be followed for hundreds of metres in an E N E - W S W direction. This shear zone contains a few asymmetric granulitic boudins indicating a dextral sense of shear, similar to the sense of shear observed in the minor shear zones in northwestern HolsnCy. Shear criteria are nevertheless relatively rare, due to an extensive greenschistfacies overprint. The dextral sense of shear is roughly compatible with the early amphibolitic top-to-the-SE phase Sd-1 described by Schmid et al. (1998) in the Rossland Shear Zone.
fully compatible with the top-to-the-NE shear sense on major shear zones.
186
H. RAIMBOURG ET AL.
Fig. 9. Map of amphibolitic deformation on northwestern Holsn0y, Rossland Shear Zone and north of Meland Nappe.
Discussion Geodynamic interpretation of the northwestern HolsnOy kinematicsgeometrical model Our field study of eclogitic shear zones reveals top-to-the-NE shear deformation. The deformation in the amphibolitic field is on average top-to-the-SE (first amphibolitic deformation (Sd-1) in Schmid et al. 1998), with a nevertheless much larger scatter than the eclogitic data. Despite these variations, there is a continuum of deformation from eclogitic to amphibolitic conditions, which leads us to interpret the eclogitic deformation as the first recorded stage of exhumation for the granulite unit.
A collisional eclogitic deformation. The history of the Caledonian orogenesis is traditionally decomposed in a collisional stage in the late Ordovician-Silurian, followed by an extensional stage starting from the Devonian. Fossen (1992) states that extension started around 400Ma with kinematics reversal along the basal decollement zone of the Caledonian nappes. As noted earlier (see 'geological setting'), there are controversies on the age of eclogite metamorphism, either 460 or 420 Ma. Whatever the age, the shortest time span of 20 Ma between all eclogite ages and the start of extensional tectonics (syn- or post-collisional) shows that the eclogitic deformation studied on HolsnCy is coeval with the collisional phase. The presence of eclogite-facies metamorphism as young as 405 + 5 Ma (Root et al. 2001; Root et al. in press) or ultra-high-pressure metamorphism
dated at 401.6 _+ 1.6 Ma (Carswell et al. 2003) or at 402 _+ 2 Ma (Tucker et al. in press) in the WGR also illustrates that while Holsn0y started to exhume, the whole orogen was experiencing convergence. HolsnOy eclogitic deformation within the Caledonides and Bergen Arcs kinematic framework. The general structure of the Caledonian orogen is thought to result from the NW-dipping subduction of Baltica under Laurentia (Krogh 1977). This subduction geometry is inferred from nappe-thrust kinematics and from variations in the degree of metamorphism in the WGR (Griffin et al. 1985). The nappes of western Norway were thrusted onto the Baltic shield toward the SE. Furthermore, the difference between the maximum burial (more than 100 km) for coesite-bearing units in the NW of the Western Gneiss Region (Smith 1984; Wain 1997; Carswell et al. 2001) and the much shallower burial (pressure peak of 13kbars, c. 40 kin) in the SE of the WGR (eclogites in the vicinity of the Solund basin; Chauvet et al. 1992; Hacker et al. 2003 is roughly in agreement with the SE-trending metamorphic gradient defined by Griffin et al. (1985). This direction is inferred to be parallel to the thrusting direction. On a more regional scale, other units of the Bergen Arcs were deformed during the Caledonian orogenesis. Within the highly sheared Major Bergen Arc, a garnet-amphibole micaschist recorded a top-to-the-SE sense of shear, corresponding to Caledonian contractional deformation (Wennberg 1996). The contractional phase in the Ulriken Gneisses and its psammitic cover is recorded in sheath-like folds and S-C
KINEMATICS OF ECLOGITE-FACIES DEFORMATION structures, both indicating thrusting toward the east (Fossen 1988). In the Oygarden Gneisses, structures indicating top-to-the-E sense of shear developed under upper greenschist-facies during the Caledonian collision (Fossen & Rykkelid 1990). Caledonian contractional structures in southwestern Norway are thus characterized by sense of shear and transport directions top-to-the-SE in the SE of the Caledonian nappes, and top-tothe-E in the Bergen Arcs (Fossen 1992). This sense of shear is roughly in agreement with the top-to-the-NE movement shown by eclogitic shear zones on northwestern Holsnoy. LateCaledonian extensional shear zones such as BASZ (Wennberg 1996) and the formation of the arcuate structure of the Bergen Arcs may have caused the slight misorientation.
Post-eclogitic rotation of northwestern HolsnOy To place the eclogitic deformation in northwestern HolsnCy within the overall Caledonian collision geometry, we also need to assess the
187
rotation that affected the granulite unit during its exhumation. In their tectonic model of Caledonian orogenesis, based on the south Norwegian Caledonides, Andersen et al. (1991) proposed that the orogenic extensional collapse was initiated by the detachment of the subducted lithospheric mantle. This resulted in isostatic rebound and eduction of deeply buried crust. Geometrically, during exhumation, the educted crust undergoes top-to-the-SE rotation (clockwise when looking NE in a N W - S E section of the subduction zone, see Fig. 10a). This general top-to-the-SE rotation, related at large scale to the NW-dipping subduction may, in detail, be expressed as top-to-the-E rotation in the Bergen Arcs, where tectonic features are slightly misorientated with respect to general geometry of the orogen (see above the distribution of shear senses, as described by Fossen (1992)). In addition, some rotations may stem from lateorogenic extension. The mode II extension proposed by Fossen (1992, 2000), is made through large and deeply-rooted normal shear zones/ faults, among which is the Bergen Arc Shear Zone. Normal movements on these listric faults cause rotations: large normal displacements
Fig. 10. (a) Model for 'top-to-the-SE rotation': the present geometry is rotated clockwise about a horizontal NE trending axis, when looking to the NE. (b) Restoration of Holsn~y in its original position within the Caledonian slab. The rotation shown on the stereographic plot corresponds to a top-to-the-W 60* rotation about a NS axis. As a result, eclogite-facies shear zones, oriented in their present position (direction: N120, dip: 20NE) with normal-dextral movement, are rotated into new position (direction: N20, dip: 50W) as reverse-dextral shear zones, corresponding to their former position in the Caledonian subducting slab.
188
H. RAIMBOURG ET AL.
along the WSW-dipping BASZ, on the eastern border of the Bergen Arc, result in top-to-ENE rotation of the nappe pile. Therefore, both processes proposed above lead roughly to topto-the-E rotation of the Lindfis nappe after eclogitic metamorphism. To restore the eclogitic shear zones to their original position at depth, within the Caledonian slab, we rotate them counterclockwise with respect to a N - S horizontal rotation axis (Fig. 10b). After about 20 ~ rotation, shear zones dip toward the W N W instead of the NE, and their normal sense of displacement becomes reverse (Fig. 10b shows the situation after an arbitrary rotation of 60~ Both the restored geometry (c. W N W dipping) of the shear zones and their reverse sense of shear are thus much more compatible with the general frame of the NW-dipping subduction.
Interpretation of eclogitic deformation.
As a result of this geometric reconstruction, we propose the following scheme for the history of crustal units affected by H P - L T metamorphism (Fig. 11). The crust, whose lower part is relatively rigid and coherent, is dragged down by the dense lithospheric mantle as the upper part of the subducting slab. During burial, P-T changes make minerals metastable and,
depending on the supply of fluids, a fraction of this crust is transformed into weaker and denser eclogite. Deformation is localized in eclogitized parts that constitute shear zones setting apart units of undeformed granulite. The crust is cut out in nappes, which can start their way up to the surface by thrusting toward the east and piling up onto each other. Such an interpretation also implies that a normal shear zone separates the rigid buttress (Laurentia) from the exhuming units within the continental accretionary wedge (see arrows along the upper boundary of the subduction wedge in Fig. 11). To our knowledge, there is no evidence for such a zone in the Bergen Arcs. An easy but not completely satisfactory explanation for this absence is that there is not a single but several units stacked in the subduction wedge. The units affected by the normal sense of shear may not have been exhumed up to the surface, unlike units affected by reverse sense deformation. Another reason for the lack of a large normal shear zone may be that the extent of reverse sense deformation within the subduction wedge is larger than the extent of normal sense deformation: reverse movements are caused by exhumation and by the displacement of the subducting slab relative to units in the subduction wedge, whereas normal movements are caused by exhumation only.
Fig. 11. Interpretative sketch of processes occurring in the buried crust. The zoom shows large eclogitic shear zones wrapping around decoupled crustal units, while eclogitic reactions propagate into untransformed crust through fractures and shear zones distributed in two main sets of orientation.
KINEMATICS OF ECLOGITE-FACIES DEFORMATION CoIlisional exhumation models Strain localization. Collisional exhumation models can be described in terms of two endmembers, penetrative corner/wedge flow models (Jischke 1975; England & Holland 1979; Cloos 1982; Shreve & Cloos 1986; Mancktelow 1995) and localizing rigid slab models (Chemenda et al. 1995). The two types of model differ first by the spatial distribution of strain (see review by Platt 1986). In wedge/corner flow models strain varies smoothly within exhumed units, whereas in Chemenda's models strain is extremely localized into two shear zones, one normal at the top and the other one reverse at the bottom, enabling a unit of undeformed crust to reach the surface after its burial. The distribution of eclogitic strain on HolsnCy is highly heterogeneous at the kilometre scale. Undeformed kmscale crustal units (i.e. untransformed granulite) are separated by highly strained 100m-scale bands (i.e. large eclogitic shear zones). This heterogeneity must be accounted for in modelling kin-scale structures and phenomena, such as subduction channel or wedge dynamics. In contrast, an average rheology that smoothes such strain-localization effects is relevant to model geodynamics of the whole orogeny. A delicate balance between shear stresses and buoyancy forces. The two end-member models also deal differently with the way shear forces are handled in exhumation processes. In Chemenda's models, exhumation is only caused by the positive buoyancy of crustal units dragged into the denser mantle. The presence of a weak decoupling layer between rigid crustal units and underlying mantle, in the sinking slab, decides whether the ascent of buoyant crust is possible or not. In corner/flow models, shear forces alone can cause upward motion: if the subduction channel that contains the subducting crust and its sedimentary cover is narrowing, the downward circulation created by basal shear stresses is balanced under certain conditions by upward flow. Despite this difference, the concept of subduction channel also assumes partial decoupling of the crustal units and the lithospheric mantle within the subducting slab. Decoupling being partially controlled by the mechanical behaviour of the buried crust, a question is to what extent a change in crustal rheology affects the processes in subduction and collision zones. Looking at complete P-T loops of a crustal unit, from near the surface to the mantle and back to the surface, a decrease in viscosity has a two-fold consequence. At depth, it increases the
189
decoupling of the crust from the downgoing lithospheric mantle and therefore it enhances its capacity to exhume. On the other hand, early exhumation of buoyant crust may prevent a large fraction of the crust from reaching and equilibrating at high pressures. The balance between both effects could only be analysed through the study of metamorphic terranes at the scale of the whole orogen, which is by far beyond the scope of this paper. In addition, on HolsnCy we do not deal with permanent rheological changes but with a change in viscosity that happens only when the crust reaches high pressures. In such circumstances, the consequences of eclogitization are quite straightforward: the rigid granulite is dragged down until it gets partially eclogitized; the eclogitized zones, weaker than the granulite, act as decoupling surfaces, which enable granulitic slivers to start their way up. Therefore, in terms of rheology, eclogitization unambiguously favours exhumation and is the first necessary step toward exhumation. An important question still to be answered about this rheological weakening is whether it is induced by hydration and grain size reduction only or also by the mineralogical change. The second major physical consequence of eclogitization is the density change, which can be up to 15% (Austrheim 1987). As a result, a completely eclogitized lower crust becomes denser than the surrounding mantle (Le Pichon et al. 1992, 1997; Dewey et al. 1993), annihilating the buoyancy force, which in all models is a major driving force to exhumation, if not the only one. In terms of exhumation, eclogitization has thus two opposite consequences. Eclogitization is an evolutionary process, in time - that is, the set of reactions progresses continuously between transformed and untransformed states - and in space - that is, some parts of the granulitic unit are still pristine while others are completely reacted. The balance between viscosity decrease and density increase, respectively promoting and impeding exhumation, depends on the way metamorphic reactions progress and propagate. The eclogitic shear zones cutting through preserved granulite, as observed on Holsn0y as well as in Flatraket (Krabbendam et al. 2000; Wain et al. 2001), enable mechanical decoupling without large density increase when averaged over the whole granulitic unit. One may imagine another situation, where very large fluid advection would induce much more pervasise eclogitization. There, the associated density increase would lead to the further sinking of eclogitized crust. Of course, such a situation is purely
190
H. RAIMBOURG ET AL.
hypothetical, since what we observe is only what has finally been exhumed; therefore, only a comprehensive mechanical modelling of eclogitization propagation could help to understand deep processes affecting rocks that finally come back up, as well those that allegedly disappear in the mantle.
Conclusions The northern part of Holsncy Island is a granulitic unit that experienced H P - L T metamorphism during the Caledonian orogenesis. This metamorphism heterogenously affected a granulite-facies protolith terrane, resulting in the juxtaposition of well-preserved granulitic zones and eclogitic zones. The eclogitic zones range from 10 cm wide reaction zones along fractures, involving only fluid diffusion from the fracture and little deformation, to eclogitized zones that are strongly deformed, forming shear zones of size ranging from 1 cm to 100 m. The comparison of P-T-t paths of the rocks on Holsncy with paths of rocks in the surrounding areas or in the W G R indicates that exhumation of the northern Holsn0y unit occurred while some close units were still being buried to large depths. A detailed study of eclogite-facies deformation in large shear zones (in particular that of asymmetrically deformed objects such as granulitic boudins or coronas) shows a consistent pattern of normal-dextral shear zones with topto-the-E sense of shear throughout northwestern Holsncy. In little transformed zones, a second set of minor shear zones, showing n o r m a l sinistral sense of shear, mimics a bookshelf mechanism. A small amount of amphibolite-facies retrogression occurs in thin shear zones crosscutting both granulite and eclogitic shear zones. The deformation in these amphibolitic shear zones reproduces roughly the eclogitic kinematic pattern, showing that eclogitic deformation recorded on Holsnoy corresponds to the first stages of exhumation. The Caledonian orogenesis resulted from the subduction of Baltica under Laurentia, in a N W - S E convergent context (Mckerrow et al. 1991; Torsvik et al. 1996). When replaced in a geodynamical context of NW-dipping subduction, the eclogitic deformation pattern can be interpreted as deep thrusting and piling-up of several crustal slivers. We thus propose the following scheme for the history of crustal units affected by H P - L T metamorphism. The crust, whose lower part was relatively rigid, was dragged down by the dense lithospheric mantle as the upper part of the
subducting slab. During burial, P-T changes made the protolith granulite-facies minerals metastable and, depending on the supply of fluids, a fraction of the crust was transformed into weaker and denser eclogite. Deformation was localized in eclogitized shear zones setting apart units of undeformed granulite. The crust was cut out into nappes that started their way up to the surface by buoyant ascent, by thrusting onto each other.
References ANDERSEN, T. B., JAMTVEIT, B., DEWEY, J. F. & SWENSSON, E. 1991. Subduction and eduction of continental crust: major mechanisms during continent-continent collision and orogenic extensional collapse, a model based on the south Norwegian Caledonides. Terra Nova, 3, 303-310. AUSTRHEIM, H. 1987. Eclogitization of lower crustal granulites by fluid migration through shear zones. Earth and Planetary Science Letters, 81, 221-232. AUSTRHEIM,H. 1990a. Fluid induced processes in the lower crust as evidenced by Caledonian eclogitization of Precambrian granulites, Bergen Arcs, Western Norway. PhD thesis, University of Oslo. AUSTRHEIM, H. 1990b. The granulite-eclogite facies transition: A comparison of experimental work and a natural occurrence in the Bergen Arcs, western Norway. Lithos, 25, 163-169. AVSTRHEIM, H. 1994. Eclogitization of the deep crust in continent collision zones. Comptes-rendus de l'Acadimie des Sciences de Paris, 319, 761-774. AUSTRHEIM, H. & GRIFFIN, W. L. 1985. Shear deformation and eclogite formation within granulitefacies anorthosites of the Bergen Arcs, Western Norway. Chemical Geology, 50, 267-281. AVSTRUEIM, H. & MORK, M. B. E. 1988. The lower continental crust of the Caledonian mountain chain: evidence from former deep crustal sections in western Norway. Norges Geologiske UndersOkelse, Special publication, 3, 102-113. AUSTRHEIM, H. & ENGVIK, A. K. 1995. Fluid transport, deformation and metamorphism at depth in a collision zone. In: JAMTVEIT, B. & YARDLEY, B. W. D. (eds) Fluid Flow and Transport in Rocks: Mechanisms and Effects. Chapman and Hall, London, 123-137. AUSTRHEIM, H., ERAMBERT, M. & ENGVlK, A. K. 1996. Processing of crust in the root of the Caledonian continental collision zone: the role of eclogitization. Tectonophysics, 273, 129-153. BASCOU, J., TOMMASI, A. & MAINPRICE, D. 2002. Plastic deformation and development of clinopyroxene lattice preferred orientations in eclogites. Geology, 24, 1357-1368. BINGEN, B., DAVIS, G. A. & AUSTRHEIM, H. 1998. Zircon growth during fluid induced Caledonian/ Scandian eclogite-facies metamorphism of the Lindfis Nappe, Caledonides of W Norway. Mineralogical Magazine, 62A, 161 - 162. BINGEN, B., DAVIS, W. J. & AUSTRHEIM, H. 2001. Zircon U-Pb geochronology in the Bergen arc
KINEMATICS OF ECLOGITE-FACIES DEFORMATION
191
eclogites and their Proterozoic protoliths, and implications for the pre-Scandian evolution of the Caledonides in western Norway. Geological Society of America Bulletin, 113, 640-649. BIRTEL, S., ALTENBERGER, U. & PASSCHIER, C. W. 1998. Polyphase structural and metamorphic evolution of the Rossland shear zone at Holsn0y, a ductile nappe boundary in the Middle Allochton of the Norwegian Caledonides. Zentralblatt fiir
CHOPIN, C. 1984. Coesite and pure pyrope in highgrade blueschists of the western Alps: a first record and some consequences. Contributions to Mineralogy and Petrology, 86, 107-118. CHOPIN, C. • SCHERTL, H. P. 2000. The UHP Unit in the Dora-Maira massif, western Alps. In: ERNST, W. G. & LIOU, J. G. (eds) Ultra-High
Mineralogie, Geologie und Paliiontologie, Teil I,
Society of America, Boulder, 133-148. CLOOS, M. 1982. Flow melanges: Numerical modelling and geologic constraints on their origin in the Franciscan complex, California. Geological Society of America Bulletin, 93, 330-345. COHEN, A. S., O'NIONS, R. K., SIEGENTHALER, R. & GRIFFIN, W. L. 1988. Chronology of the pressuretemperature history recorded by a granulite terrain. Contributions to Mineralogy and Petrology, 99, 303-311. DEWEY, J. F., RYAN, P. D. & ANDERSEN, T. B. 1993. Orogenic uplift and collapse, crustal thickness, fabrics and metamorphic phase changes: the role of eclogites. In: PRICHARD,H. M., ALABASTER,T., HARRIS, N. B. W. & NEARY, C. R. (eds) Magmatic Processes and Plate Tectonics. Geological Society of London, Special Publications, 76, 325-343. ENGLAND, P. C. & HOLLAND, T. J. B. 1979. Archimedes and the Tauern eclogites: the role of buoyancy in the preservation of exotic eclogite blocks. Earth and Planetary Science Letters, 44, 287-294. ERNST, W. G. & LIOU, J. G. 2000. Overview of UHP metamorphism and tectonics in well-studied collisional orogens. In: ERNST, W. G. & Llou, J. G. (eds) Ultra-High Pressure Metamorphism and
1-2, 19-44. BOUNDY, T. M., FOUNTAIN, D. M. & AUSTRHEIM, H. 1992. Structural development and petrofabrics of eclogite facies shear zones, Bergen Arcs, Western Norway: implications for deep crustal deformational processes. Journal of Metamorphic Geology, 10, 127-146. BOUNDY, T. M., ESSENE, E. J., HALL, C. M., AUSTRHEIM, H. & HALLIDAY, A. N. 1996. Rapid exhumation of lower crust during continentcontinent collision and late extension: Evidence from 40Ar/39Ar incremental heating of hornblendes and muscovites, Caledonian orogen, western Norway. Geological Society of America Bulletin, 108, 1425-1437. BOUNDY, T. M., MEZOER, K. & ESSENE, E. J. 1997a. Temporal and tectonic evolution of the granuliteeclogite association of the Bergen Arcs, western Norway. Lithos, 39, 159-178. BOUNDY, T. M., HALL, C. M., LI, G., ESSENE, E. J. 8r HALLIDAY, A. N. 1997b. Fine-scale isotopic heterogeneities and fluids in the deep crust; a 40Ar/39Ar laser ablation and TEM study of muscovites from a granulite-eclogite transition zone. Earth and Planetary Science Letters, 148, 223-242. BUROV, E., JOLIVET, L., LE POURHIET, L. & POLIAKOV, A. 2001. A thennomechanical model of exhumation of high pressure (HP) and ultrahigh pressure (UHP) metamorphic rocks in Alpine-type collision belts. Tectonophysics, 342, 113-136. CARSWELL, T., CUTHBERT, S., O'BRIEN, P. 8r TUCKER, R. 2001. Eclogite evolution and the timing of ultra-high pressure metamorphism in western Norway. EUG XI, Strasbourg, France. Journal of Conference Abstracts, 6, abstract 348. CARSWELL, D. A., TUCKER, R. D., O'BRIEN, P. J. 8r KROGH, T. E. 2003. Coesite micro-inclusions and the U/Pb age of zircons from the Hareidland Eclogite in the Western Gneiss Region of Norway. Lithos, 67, 181 - 190. CHAUVET, A., KIENAST, J. R., PINARDON, J. L. & BRUNEL, M. 1992. Petrological constraints and PT path of Devonian collapse tectonics within the Scandian mountain belt (Western Gneiss Region, Norway). Journal of the Geophysical Society, London, 149, 383-400. CHEMENDA,A. I., MATTAUER, M., MALAVIEILLE,J. 8r BOKUN, A. N. 1995. A mechanism for syn-collision rock exhumation and associated normal faulting: results from physical modelling. Earth and Planetary Science Letters, 132, 225-232.
Pressure Metamorphism and Geodynamics in Collision-Type Orogenic Belts. Geological
Geodynamics in Collision-Type Orogenic Belts. Geological Society of America, Boulder, 3-19. FOSSEN, H. 1988. The Ulriken gneiss complex and the Rundemanen Formation: a basement-cover relationship in the Bergen Arcs, West Norway. Norges Geologiske UndersOkelse Bulletin, 412, 67-86. FOSSEN, H. 1992. The role of extensional tectonics in the Caledonides of south Norway. Journal of Structural Geology, 14, 1033-1046. FOSSEN, H. 2000. Extensional tectonics in the Caledonides: Synorogenic or postorogenic? Tectonics, 19, 213-224. FOSSEN, H. & RYKKELID, E. 1990. Shear zone structures in the Oygarden area, West Norway. Tectonophysics, 174, 385-397. GRIFFIN, W. L., AUSTRHEIM, H., BRASTAD, K., BRYHNI, I., KRILL, A. G., KROGH, E. J., MORK, M. B. E., QVALE, H. & TORUDBAKKEN, B. 1985. High-pressure metamorphism in the Scandinavian Caledonides. In: GEE, D. G. & STURT, B. A. (eds) The Caledonide Orogen-Scandinavia and Related Areas (Vol. 2). John Wiley & Sons Ltd, Chichester, UK, 783-801. HACKER, B. R., ANDERSEN, T. B., ROOT, D. B., MEHL, L., MATTINSON, J. M. & WOODEN, J. L. 2003. Exhumation of high-pressure rocks beneath the Solund Basin, Western Gneiss Region of Norway. Journal of Metamorphic Geology, 21, 613-629.
192
H. RAIMBOURG ETAL.
JAMTVEIT, B., BUCHER-NURMINEN,K. & AUSTRHEIM, H. 1990. Fluid controlled eclogitization of granulites in deep crustal shear zones, Bergen arcs, Western Norway. Contributions to Mineralogy and Petrology, 104, 184-193. JISCHKE, M. C. 1975. On the dynamics of descending lithospheric plates and slip zones. Journal of Geophysical Research, 80, 4809-4813. JOLIVET, L., FACCENNA, C., GOFFE, B., BUROV, E. & AGARD, P. 2003. Subduction tectonics and exhumation of high-pressure metamorphic rocks in the Mediterranean orogens. American Journal of Science, 303, 353-409. KOLDERUP, C. F. & KOLDERUP, N. H. 1940. Geology of the Bergen Arc System. Bergens Museum Skrifter, 137 p. KRABBENDAM, M., WAIN, A. & ANDERSEN, T. B. 2000. Pre-Caledonian granulite and gabbro enclaves in the Western Gneiss Region, Norway: Indications of incomplete transition at high pressure. Geological Magazine, 137, 235-255. KROGH, E. J. 1977. Evidence of Precambrian continentcontinent collision in Western Norway. Nature, 267, 17-20. KOHN, A. 2002. The influence of fluid on the granulite to eclogite and amphibolite facies transition: a study in the anorthositic rocks from the Lind~ts Nappe, Bergen Arcs, West Norway. PhD thesis, University of Oslo. LABROUSSE, L. 2001. L'exhumation des roches m~tamorphiques de trks haute pression: le cas des Calddonides de Norvkge. Thrse de l'Universit~ Pierre et Marie Curie, Paris. LE PICHON, X., FOURNIER, M. & JOLIVET, L. 1992. Kinematics, topography, shortening and extrusion in the India-Eurasia collision. Tectonics, 11, 1085-1098. LE PICHON,X., HENRY, P. & GOFFE, B. 1997. Uplift of Tibet: from eclogites to granulites - implications for the Andean Plateau and the Variscan Belt. Tectonophysics, 273, 57-76. MANCKTELOW, N. S. 1995. Nonlithostatic pressure during sediment subduction and the development and exhumation of high pressure metamorphic rocks. Journal of Geophysical Research, 100, 571-583. MATTEY, D., JACKSON, D. H., HARRIS, B. W. & KELLY, S. 1994. Isotopic constraints on fluid infiltration from an eclogite facies shear zone, Holsn0y, Norway. Journal of Metamorphic Geology, 12, 311-325. MCKERROW, W. S., DEWEY, J. F. & SCOTESE, C. R. 1991. The Ordovician and Silurian development of the Iapetus ocean. Special Papers in Palaeontology, 44, 165-178. PASSCHIER, C. W. & TROUW, R. A. J. 1998. Microtectonics. 2nd ed. Springer-Verlag, Berlin, Heidelberg, 289 p. PLATT, J. P. 1986. Dynamics of orogenic wedges and the uplift of high-pressure metamorphic rocks. Geological Society of America Bulletin, 97, 1037-1053. PLATT, J. P. 1993. Exhumation of high-pressure rocks: a review of concept and processes. Terra Nova, 5, 119-133.
RAGNHILDSVEIT,J. & HELLIKSEN,D. 1997. Geologisk kart over Norge, begrunnskart Bergen, 1:250000. Norges Geologiske UndersCkelse. REY, P., FOUNTAIN, D. M. & BOUNDY, T. M. 1999. Kinematic indicator in eclogite-facies shear zones. In: SNOKE, A. W., TULLIS, J. A. and TODD, V. R. (eds) Fault-Related Rocks - A Photographic Atlas. Princeton University Press, 629 pp. RORERTS, D. & GEE, D. G. 1985. An introduction to the structure of the Scandinavian Caledonides. In: GEE, D. G. & STURT, B. A. (eds) The Caledonide Orogen: Scandinavia and Related Areas (Vol. 1). John Wiley, Chichester, 55-68. ROOT, D. B., HACKER, B. R., MATTINSON, J. M. & WOODEN, J. L. 2001. U/Pb zircon geochronology of High- and Ultrahigh-pressure eclogites of the Western Gneiss Region, Norway. l lth Annual V. M. Goldschmidt Conference, Houston, abstract 3883. ROOT, D. B., HACKER, B. R., MATTINSON, J. M. & WOODEN, J. L. 2004 Young age and rapid exhumation of Norwegian Ultrahigh-pressure rocks; an ion microprobe and chemical abrasion study. Earth and Planetary Science Letters, in press. SCHMID, R., ALTENBERGER, U. & OBERHANSLI, R. 1998. Polyphase tectonometamorphic evolution of the northwestern Lindfis Nappe on HolsnCy, Bergen Arcs, Caledonides, S-W Norway. Zentralblatt fiir Mineralogie, Geologie und Paliiontologie, Teil I, 1-2, 1-18. SHREVE, R. L. & CLOOS, M. 1986. Dynamics of sediment subduction, melange formation, and prism accretion. Journal of Geophysical Research, 91, 10.229-10.245. SMITH, D. C. 1984. Coesite in clinopyroxene in the Caledonides and its implications for geodynamics. Nature, 310, 641-644. TORSVIK, T. H. & EIDE, E. A. 1998. NGUGEOCHRO: Database and analysis for Norwegian isotope geochronology. NGU Open File Report 98.003, 54. TORSVIK, T. H., SMETHURST, M. A. et al. 1996. Continental break-up and collision in the Neoproterozoic and Palaeozoic - a tale of Baltica and Laurentia. Earth Science Reviews, 40, 229-258. TUCKER, R. D., ROBINSON, P., GEE, D. G., THORSNES, T., KROGH, T. E., NORDGULEN, 0. & BICKFORD, M. E. Thrusting and extension in the Scandian hinterland, Norway: New U-Pb ages and tectonostratigraphic evidence. American Journal of Science, in press. WAIN, A. 1997. New evidence for coesite in eclogite and gneisses: Defining an ultra-high pressure province in the Western Gneiss region of Norway. Geology, 25, 927-930. WAIN, A., WATERS, D. J. & AUSTRHEIM, H. 2001. Metastability of granulites and processes of eclogitisation in the UHP region of western Norway. Journal of Metamorphic Geology, 19, 607-623. WENNBERG, O. P. 1996. Superimposed fabrics due to reversal of shear sense: an example from the Bergen Arc Shear Zone, western Norway. Journal of Structural Geology, 18, 871-889.
Steep extrusion of late Archaean granulites in the Northern Marginal Zone, Zimbabwe: evidence for secular change in orogenic style T H O M A S G. B L E N K I N S O P 1 & A L E X A N D E R F. M. KISTERS 2
1School of Earth Sciences, James Cook University, Townsville, QLD 4811, Australia (e-mail: Thomas. Blenkinsop @jcu. edu. au) 2Department of Geology, University of Stellenbosch, Private Bag X1, Matieland 7602, South Africa Abstract: The Northern Marginal Zone (NMZ) of the Limpopo Belt in southern Zimbabwe
is an essentially plutonic terrain, now consisting of granulite-gneisses, that was assembled by extensive magmatic under- and intraplating between c. 2.7 and 2.6 Ga during late Archaean crustal shortening. In this study, we present the first evidence for regional-scale normal shear in the NMZ. Normal shear has been identified along an up to 500 m wide, steeply dipping belt of mylonites, the Mtilikwe shear zone. Geochronological and mineralogicaltextural data indicate that normal shear occurred coeval with crustal shortening that was accommodated along gently dipping thrust zones. The bulk strain pattern consisted of horizontal shortening, vertical stretching, and subvertical extrusion of a crustal segment. Horizontal shortening persisted throughout the orogenic evolution in the NMZ and no features typical of gravitational collapse in modern orogens are seen. This could imply that there was never major overthickening of the crust by thrusting, which may be a characteristic feature of Archaean orogeny.
The long-standing debate about the similarity between Archaean and modern tectonics has mostly focused on the classic granite-greenstone terranes of the Archaean cratons (e.g. Kr6ner 1981; de Wit 1998; Hamilton 2004), which exhibit a distinctive tectonic style (Choukroune et al. 1995). However, Archaean gneissic terranes, such as the Limpopo belt in southern Africa, may be more useful in this discussion because comparisons can be readily made with younger orogenic belts (e.g. Marshak 1999). The rationale for possible changes in tectonic style from the Archaean to the present is that the Earth contained more thermal energy due to retention of primordial heat and higher concentration of radioactive elements. This may have elevated crustal geotherms, although this point is debated. The concept of gravitational collapse is increasingly seen as an integral part of orogenesis (Rey et al. 2001). Gravitational collapse may occur in the late- to post-collisional evolution of orogens in response to tectonic overthickening during collision and/or the delamination or convective removal of the lower parts of the thickened lithospheric keel (e.g. Dewey 1988; England & Houseman 1988, 1989). Postconvergence gravitational collapse would be
facilitated by higher geothermal gradients in the crust, and it may also have been enhanced in the Archaean by delamination due to more vigorous mantle circulation (Sandiford 1989a). Fabrics that result from gravitational collapse of midcrustal levels are commonly characterized by flat-lying foliations, recumbent folds and low-angle thrusts that accommodate the vertical shortening and lateral spreading of the crust (e.g. Sandiford 1989b). The component of vertical shortening in the thickened orogenic belt may be enhanced by the catastrophic delamination or convective removal of the lithospheric keel during which the overlying crust rebounds, undergoing further vertical shortening (e.g. Bird 1979; Houseman et al. 1981; Platt & England 1994). Given the higher heat flow in the Archaean, Marshak (1999) suggested that Archaean orogenic belts may be narrower and lower than Phanerozoic belts for a given amount of horizontal shortening, and that they might contain relatively wider belts of plastically deformed rock with shallowly dipping extensional fabrics formed during orogenic collapse. The Northern Marginal Zone (NMZ) in southern Zimbabwe is a late Archaean, granulitegneiss belt that borders the lower-grade metamorphic granite-greenstone terrain of the
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics:from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 193-204. 0305-8719/05/$15.00
9 The Geological Society of London 2005.
194
T.G. BLENKINSOP & A. F. M. KISTERS
Zimbabwe Craton to the north (James 1975) (Fig. 1). This narrow belt of high-grade metamorphic rocks records extensive midcrustal plutonism, high-grade metamorphism and associated contractional deformation between c. 2.7 and 2.6 Ga (e.g. Ridley 1992; Mkweli et al. 1995; Rollinson & Blenkinsop 1995; Berger et al. 1995; Kamber et al. 1996). Peakmetamorphic P - T conditions have reached 800MPa and 8 0 0 - 8 5 0 ~ (Rollinson 1989; Kamber & Biino 1995), with an anticlockwise P - T path. Given the similar present-day crustal thicknesses of the NMZ and the Zimbabwe Craton (c. 35-40 km; Stuart & Zengeni 1989; Nguuri et al. 2001), the maximum possible late Archaean crustal thickness may have been 60 km (Rollinson 1989; Kamber et al. 1996; Blenkinsop et al. 2005). The boundary between the NMZ and the Zimbabwe Craton is marked by a thrust zone, the North Limpopo Thrust Zone (Blenkinsop et al. 1995) (Fig. 1), and earlier models of crustal thickening have mainly invoked tectonic stacking during continental collision or terrane accretion and top-tothe-N directed thrusting of rocks of the NMZ onto the Zimbabwe Craton (e.g. Roering et al. 1992; de Wit et al. 1992; Rollinson 1993). More recent works, however, allow for interpreting the crustal thickening of the NMZ to be the result of a magmatic under- and intraplating by
voluminous charnoenderbitic plutons, coinciding with the main phase of craton-wide plutonism and stabilization of the Zimbabwe Craton in the outgoing Archaean (Ridley 1992; Berger et al. 1995; Rollinson & Blenkinsop 1995). The NMZ is an ideal locality to investigate possible secular changes in orogenic style related to elevated geotherms in the Archaean. No evidence has been presented that the thickened and hot crustal section of the NMZ has undergone gravitational collapse. Gneisses in the NMZ are characterized by mainly steeply dipping gneissose structure and downdip lineations, and late Archaean shear zones commonly record top-to-the-N reverse sense of movement, all consistent with the main phase of northerly vergent contractional deformation in the NMZ and the adjacent Zimbabwe Craton at c. 2.6 Ga (Rollinson & Blenkinsop 1995). This raises the question of how the NMZ escaped widespread gravitational collapse, which seems inevitable for the theologically weakened and thickened midcrustal levels. In this contribution we describe for the first time a regional-scale normal sense shear zone (the Mtlikwe shear zone) in the midcrustal rocks of the NMZ. We present evidence that normal shearing occurred under high-grade metamorphic conditions and largely coeval with the main phase of crustal shortening and
Fig. 1. The high-grade metamorphic Limpopo Belt in southern Africa, situated between the Archaean, mainly low-grade granite-greenstone terrains of the Zimbabwe and Kaapvaal Cratons. The location of the study area in the Northern Marginal Zone is shown in the white box: the box in the inset shows the location of the map.
EXTRUSION OF LATE ARCHAEAN GRANULITES thickening. Significantly, the belt of steeply dipping normal-sense mylonites is contained within the regional structural grain of the NMZ; that is, it is parallel to the regional gneissose structure and lineation patterns that have formed during the main contractional phase at c. 2.6 Ga. Normal sense shearing is only observed through the detailed observation of kinematic indicators. This suggests that regionalscale normal shear in the NMZ may be more common than hitherto recognized. The recognition of high-grade metamorphic normal shear zones in the NMZ bears on the exhumation history and mechanisms of the NMZ, the juxtaposition of the high-grade terrain against lower crustal rocks of the Zimbabwe Craton, and possibly on secular variations in tectonic style.
Regional geological setting The ENE trending NMZ forms the northern parts of the late Archaean to Mid-Proterozoic Limpopo Belt, a granulite-gneiss belt that separates the mainly low-grade metamorphic Archaean granite-greenstone terrains of the Zimbabwe Craton in the north from the Kaapvaal Craton in the south (Fig. 1). The NMZ is exposed over a strike length of c. 450 km and a width of c. 60 km in southern Zimbabwe, and comprises three main lithological components, including (1) an older (> c. 2.71 Ga) supracrustal sequence, (2) a voluminous plutonic assemblage (c. 2.712.57 Ga) made up of intrusive enderbites and charnockites, and (3) a later suite of granites and charnockites (c. 2.62-2.57 Ga), referred to as the Razi Suite (Odell 1975; Ridley 1992; Berger et al. 1995; Kamber & Biino 1995). The plutonic charnockites and enderbites, collectively referred to as the charnoenderbite suite (Berger et al. 1995), are by far the most common rocks and the NMZ represents essentially a midcrustal plutonic terrain that was assembled during a protracted period of magmatic under- and intraplating between c. 2.7 and 2.6 Ga (e.g. Ridley 1992; Berger et al. 1995; Rollinson & Blenkinsop 1995; Blenkinsop et al. 2005). Commonly rounded outlines of massive charnoenderbites are interpreted to represent the emplacement of the dry magmas as diapirs into the lower crust (Rollinson & Blenkinsop 1995). The P - T conditions determined by mineral assemblages in supracrustal enclaves from throughout the NMZ show a considerable range from 600 to 850 ~ at pressures of 400-850 MPa (Rollinson 1989; Kamber & Biino 1995). Episodes of repeated hightemperature metamorphism are related by Kamber & Biino (1995) and Kramers et al.
195
(2001) to the prolonged intrusive activity and associated heat advection. P - T results from the study area around Renco mine (Fig. 1) show a range from 600 MPa, 825 ~ near the North Limpopo thrust Zone to 820 MPa, 785 ~ 6 km south of Renco (Rollinson 1989). Various models have been proposed for the origin of the voluminous charnoenderbitic plutonism in the NMZ (Ridley 1992; Berger et al. 1995, Berger & Rollinson 1997). Ridley (1992) suggested a non-plate-tectonic setting for the charnoenderbites, invoking lower-crustal melting and magmatic accretion of the felsic magmas above a zone of mantle downwelling. Geochemical and age similarities between the charnoenderbite suite and the tonalitetrondhjemites of the Zimbabwe Craton suggest that the mid- to lower-crustal plutonic terrain exposed in the NMZ represents a deeper section of the Zimbabwe Craton (Berger et al. 1995; Rollinson & Blenkinsop 1995; Berger & Rollinson 1997). The importance of a very high geothermal gradient and extreme rate of heat generation in the NMZ that appears to be incompatible with any modern-day geological setting has been emphasized by Kramers et al. (2001). Structures in the NMZ are heterogeneously developed and the strain intensity varies from massive, largely undeformed or only weakly foliated charnoenderbites to pervasive protomylonitic and mylonitic fabrics contained in several hundred metre wide, curved high-strain zones that trend subparallel to the regional gneissose fabrics. The regional gneissic banding trends ENE, parallel to the trend of the NMZ, and dips at mainly steep angles (40-80 ~ in a southerly direction (Figs 2 and 3). A downdip mineral stretching lineation is ubiquitous in foliated rocks (Fig. 2). Protomylonites and mylonites show kinematic indicators that point to a topto-the-NNW reverse and thrust sense of movement in some of these shear zones (the medium-grade shear zones in Fig. 3) and the regional structural inventory is consistent with a bulk NNW-SSE subhorizontal shortening strain during the late Archaean (James 1975; Ridley 1992; Roering et al. 1992; Rollinson & Blenkinsop 1995; Mkweli et al. 1995; Blenkinsop et al. 1995). The contact between the high-grade gneisses of the NMZ and the Zimbabwe Craton is a more than 250 km long and several kilometre wide composite thrust zone, collectively referred to as the North Limpopo Thrust Zone (NLTZ, Blenkinsop et al. 1995) (Fig. 1). In detail, the NLTZ consists of an anastomosing network of southeasterly dipping mylonitic and protomylonitic gneisses that enclose low-strain areas in
196
T . G . B L E N K I N S O P & A. F. M. KISTERS
Fig. 2. Contoured stereoplots of poles to foliations (left, 239 data) and lineations (right, 43 data) from throughout the Northern Marginal Zone. Equal area, lower hemisphere projections. Contouring on the surface of the projection sphere; contours in multiples of a uniform distribution, starting at 1 and in intervals of 2 x multiples of a uniform distribution (Rollinson & Blenkinsop 1995).
Fig. 3. Geology of the Mtilikwe shear zone (cross-hatched) and adjacent areas in the Northern Marginal Zone of the Limpopo Belt. Partly based on Chiwara (2003). A - A ' is part of the line of section shown in Figure 10.
EXTRUSION OF LATE ARCHAEAN GRANULITES
197
which the primary magmatic textures of the igneous precursors can still be discerned. Kinematic indicators along the moderately (30-45 ~ SE-dipping NLTZ point to a top-tothe-NW reverse sense of shear, consistent with the southeasterly plunging downdip lineation in mylonites. Frei et al. (1999) have presented geochronological evidence that thrusting was diachronous along the entire length of the NLTZ and progressed from west to east with time between c. 2.65 Ga and 2.57 Ga. Syn- to late-kinematic, 2 . 6 4 - 2 . 6 G a granites of the Razi Suite constrain the timing of thrusting in the west-central portion of the NLTZ (e.g. Mkweli et al. 1995; Berger et al. 1995). An upper age bracket for thrust deformation and uplift of the NMZ along this part of the NLTZ is provided by the intrusion of the Great Dyke that has recently been constrained to 2575_+ 0.7 Ma (Oberthiir et al. 2002). This also indicates that uplift and juxtaposition of the NMZ against the low-grade granite-greenstone terrain of the Zimbabwe Craton has occurred within a very short time span, possibly as little as 10-20 Ma. The Palaeoproterozoic (c. 2 Ga) tectonism that has affected the southern parts of the NMZ (Kamber et al. 1995) is only manifest by locally developed greenschist-facies shear zones, and the structural grain preserved in the NMZ mainly records the late-Archaean tectonism between the Zimbabwe Craton and the Limpopo Belt (Blenkinsop et al. 2005). Detailed mapping of the region around the Mtilikwe shear zone is given in Chiwara (2003), which also documents the very limited extent of the Palaeoproterozoic structures.
Study methods and samples Orientated samples and structural measurements were taken throughout the known 25 km length of the Mtilikwe shear zone. The shear zone is exceptionally well exposed in the east in a river platform of the Mtilikwe river, at Rupati Pools (Figs 3 and 4a), and moderately well exposed along the length of the Gurutsime fiver and its tributaries in the west (Fig. 3). Samples were cut perpendicular to foliation and lineation, and examined by optical microscopy and SEM.
Geology of the Mtilikwe shear zone The Mtilikwe shear zone is an up-to-500 m wide east-northeasterly trending zone of protomylonites and mylonites that can be traced for 25 km along strike (Fig. 3). The belt of mylonitic
Fig. 4. The Mtilikwe shear zone: (a) Rupati Pools outcrop, showing excellent exposure and strong mylonitic foliation dipping steeply to the southeast; (b) downdip mineral lineation on foliation surface defined by quartz and feldspar; (c) Intrafolial isoclinal folds defined by garnet-rich layers.
rocks occurs 15 km to the south of the NLTZ. P b - P b step leach ages from synkinematic garnets from within the mylonites yielded a well-defined isochron indicating an age of 2601 _+ 5 Ma that was interpreted by Blenkinsop & Frei (1996) to represent the age of shearing and accompanying metamorphism. This age is within error to the main phase of thrusting recorded along the east-central parts of the
198
T.G. BLENKINSOP & A. F. M. KISTERS
NLTZ (Mkweli et al. 1995) so that the Mtilikwe shear zone was previously described as part of the regionally developed, anastomosing system of reverse and thrust zones that constitute the composite NLTZ (Blenkinsop & Frei 1996). For most of its extent, the Mtilikwe shear zone is developed in plutonic enderbites and mediumgrained granulites as well as locally developed, strongly gametiferous and sillimanite-bearing gneisses. The latter show a pronounced compositional banding on outcrop scale and are possibly of sedimentary origin (Fig. 4c). The mylonitic foliation in the Mtilikwe shear zone trends ENE and dips steeply to the SSE, parallel to the regional structural grain of the NMZ (Figs 3, 4 and 5). A prominent lineation, made up of quartz and quartz-feldspar rods, has mainly a downdip orientation in the foliation (Fig. 4b). A slight but systematic variation in the plunge of the lineation is noted along strike, changing from southerly plunges in the west to southeasterly plunges in the east of the Mtilikwe shear zone (Fig. 5). Centimetre- to decimetre-scale isoclinal, intrafolial rootless folds in the gneissic layering testify to the transposition of an earlier compositional banding or gneissic layering in probable paragneisses (Fig. 4c). The intrafolial folds plunge parallel to the downdip rodding lineation. Sheath folds were not observed. The
trend of the Mtilikwe shear zone is very closely parallel to the regional gneissic fabric, and lineations in the MSZ are parallel to the SE to S plunge of the lineations in the regional fabric (Fig. 6).
Petrology of the mylonites Protomylonites and mylonites of the Mtilikwe shear zone are composed of high-temperature mineral assemblages including quartz-plagioclase-biotite-ortho and clinopyroxene in enderbitic rocks; quartz-alkali feldspar (microcline and orthoclase)-plagioclase-biotite-ortho- and clinopyroxene in felsic granulites of probably chamockitic origin; and quartz-alkali feldspar (microcline and orthoclase)-plagioclasebiotite-garnet- sillimanite in paragneisses (Fig. 7a). Zircon, apatite, ilmenite and rutile are common accessory minerals in most samples. Texturally, the protomylonites and mylonites of the Mtilikwe shear zone are characterized by extensive dynamic recrystallization of almost all mineral components, which results in the typically finer grain size of the shear zone rocks compared to the surrounding, massive charnoenderbites. Quartz forms several centimetre long ribbons that define the mylonitic
+
Poles to foliation
Foliation
Lineation
Rupati Pools 10 km
Fig. 5. Lower hemisphere,equal area projections of mylonitic fabrics along the Mtilikwe shear zone.
EXTRUSION OF LATE ARCHAEAN GRANULITES
199
Fig. 6. Foliation trajectory map and lineations of gneissic fabrics in the study area, with the Mtilikwe shear zone shown.
foliation (Fig. 7b, c, d). Alkali feldspar and plagioclase commonly occur as augen-shaped mantled porphyroclasts (Fig. 7b, c), but they may also be pervasively recrystallized forming composite feldspar ribbons that alternate with quartz ribbons and thereby imparting a foliation-parallel compositional banding to the mylonites. Garnet appears mainly subrounded and fractured (Fig. 8). The fractures are filled by quartz, feldspar and biotite of the main mineral assemblage indicating that fracturing occurred during the high-temperature and overall ductile deformation (Fig. 8b). Orthopyroxene is only observed in enderbitic protoliths. It commonly displays a prominent undulose extinction and marginal recrystallization into smaller aggregates. The fine-grained, recrystallized orthopyroxene aggregates are locally replaced by biotite. Overall, there is little evidence of retrogression of the rocks. Minor sericite is clearly post-kinematic and can be seen to replace feldspars along cleavage planes. In summary, the extensive dynamic recrystallization of almost all mineral components and the high-grade mineral parageneses preserved in the mylonites suggest that normal shearing along the Mtilikwe shear zone has occurred
close to or slightly post-peak metamorphic granulite-facies conditions.
Kinematic indicators Macroscopic shear sense indicators along the Mtilikwe shear zone are relatively rare and are virtually restricted to relatively coarse-grained rocks such as mylonitized pegmatites. On a microscopic scale, however, shear sense indicators are common, including o'- and ~5-clasts, S-C and S-C' fabrics (Fig. 7). Twenty orientated thin sections were studied, taken along the strike extent of the Mtilikwe shear zone. Kinematic indicators in all sections consistently point to a normal, south-side down sense of shear. There was no evidence for a reverse, top-to-the-NNW sense of shear that commonly characterizes mylonites in the NMZ. A distinctive possible shear sense indicator was observed from the orientation of extension microfractures within the garnet porphyroclasts of the mylonites (Fig. 8). These have a very consistent preferred orientation relative to the mylonitic foliation, inclined at 4 0 - 5 0 ~ to the foliation towards the shortening quadrant of normal sense shearing established independently by other
200
T . G . BLENKINSOP & A. F. M. KISTERS
Fig. 7. Optical and scanning electron microscope photomicrographs of fabrics from the Mtilikwe shear zone. All sections are parallel to the lineation and perpendicular to the foliation, and have been orientated to show the same view, in which dextral shear in the photomicrographs is an extensional shear sense in the field. (a) Sillimanite fabric (dark grey, needle-like inclusions) in large garnet. Biotite (flaky mineral at the tips of the garnet), ilmenite (bright white). Matrix of quartz and K-feldspar. Rupati Pools outcrop. SEM. (b) Delta clast of feldspar in quartz ribbons, indicating dextral shear sense. Rupati Pools outcrop. Cross-polarized light. (c) Sigma clast of microperthite (Per) showing dextral shear sense, surrounded by quartz ribbons. Sample from far west end of Mtilikwe shear zone (UTM coordinates 0293790 7709090). SEM. (d) S-C fabric (lines show S and C orientations) and feldspar sigma clast showing dextral shear sense. Rupati Pools outcrop. Plane polarized light.
Fig. 8. Extension microfractures (arrows) in garnet porphyroclasts, inclined to the foliation to indicate a dextral (extensional) shear sense. Specimens from Rupati Pools outcrop. (a) Parallel alignment occurs between several porphyroclasts. Plane polarized light. (b) Microfracture fillings comprise biotite, quartz and K-feldspar. Quartz and K-feldspar define the mylonitic foliation around the garnet porphyroclast. SEM.
EXTRUSION OF LATE ARCHAEAN GRANULITES
201
Fig. 9. Microfracture orientations in garnet relative to S and C fabrics in garnetiferous mylonites from the Rupati Pools locality. The orientations of 149 microfractures were measured from over 20 garnets in a section perpendicular to the mylonitic foliation and parallel to the lineation. Shear sense was established independently from S-C fabrics, o-- and 6-porphyroclasts. Rose diagrams are moving averages in 10 windows. (a) Microfracture frequency (scale is average number of microfratures/'3. (b) Microfracture length (scale is average length, ~m/~3.
criteria (Fig. 9). A simple interpretation of these microfractures is that they are tensile fractures formed during normal simple shear.
Discussion The interpretation of the structural data presented here depends critically on whether the structures are still in approximately the same orientation in which they were formed. This can be confirmed by the observation that the Great Dyke and its satellites intruded the Northern Marginal Zone in the late Archaean, yet they are not deformed or rotated (Fig. 1; Blenkinsop et al. 2005). The significance of this observation can be extrapolated along strike to the study area because the gneissose structure in the study area has the same dip as generally observed throughout the 450 km length and 60 km width of the NMZ (Fig. 2). Geochronological constraints (Blenkinsop & Frei 1996), high-grade metamorphic mineral parageneses and deformation textures recorded in mylonites of the Mtilikwe shear zone all indicate that normal, top-to-the-S shearing occurred approximately synchronous with the late Archaean, overall contractional deformation in the NMZ at c. 2.6 Ga. Significantly, the planar and linear fabrics of the Mtilikwe shear zone are parallel to and virtually indistinguishable from the steep regional gneissose structures and downdip linear fabrics that characterize the reverse- and thrust-sense shear zones throughout
the NMZ (Figs 2, 5 and 6). The normal shear sense can only be seen at the outcrop scale in the few coarse layers, and is difficult to find in outcrop even in very well-exposed areas such as the platform in the Mtilikwe river. This means that normal-sense shear may be more widespread in the NMZ than hitherto recognized. Two important features need to be considered when discussing the syn- to late-collisional evolution of the NMZ. First, a substantial amount of crustal thickening in the NMZ was achieved by magmatic intra- and underplating that occurred over a protracted period of over 100 Ma, between c. 2.7 and 2.6 Ga (e.g. Berger et al. 1995) and late Archaean geothermal gradients were very high due to the anomalously high radiogenic heat production in the NMZ (Kramers et al. 2001). Secondly, magmatic accretion occurred during N - S crustal shortening (e.g. Ridley 1992; Rollinson & Blenkinsop 1995; Berger et al. 1995). The crustal strength must have been low throughout the evolution of the NMZ and it seems unlikely that this hot and rheologically weak crustal section could have supported large vertical loads due to tectonic thickening. Any vertical tectonic loading of the NMZ during the late Archaean NNW-directed crustal shortening was likely to have been compensated for almost instantaneously. Notably, normal sense shearing along the Mtilikwe shear zone is not a postcollisional feature, but is synchronous with the main phase of crustal thickening. Thus, significant
202
T.G. BLENKINSOP & A. F. M. KISTERS
tectonic overthickening and subsequent thermal equilibration as some of the main prerequisites for regional-scale, extensional collapse are unlikely to have occurred in the NMZ. This observation is consistent with Marshak's (1999) suggestion that Archaean orogens were probably characterized by relatively low topographic relief. While late orogenic events in modern orogens appear to involve extensional collapse and the formation of subhorizontal fabrics in the midlower crust, the latest events in the NMZ were concurrent thrusting on predominantly lowangle structures and normal faulting on steep structures. The bulk strain accommodated by the shear systems and the pervasive, subvertical fabrics and stretching lineations is NNW horizontal shortening and vertical extension. Several possibilities exist to explain this tectonic scenario. A transpressional regime with a high ratio of horizontal shortening to wrench ('pure shear dominated transpression'; Tikoff & Teyssier 1994) could account for the vertical stretch (e.g. Pelletier et al. 2002). However, there is no evidence for a wrench component to the deformation, or indeed for any orogen parallel transport. The dextral transpression seen in the Triangle Shear zone on the southern margin of the NMZ occurred 500 Ma later than the Archaean tectonics discussed here (Kamber et al. 1995). Isostatic readjustment following crustal underplating might explain the observations, but the likely locus of maximum underplating is to the south of the study area, where the highest grade metamorphic conditions were reached. The observed normal sense of shear on the Mtilikwe shear zone is the opposite of that expected for such isostatic readjustment. Horizontal gradients in vertical stretching could account for the concurrent operation of the thrusts and normal faults if the vertical stretch
was concentrated in the area between the Mtilikwe shear zone and the North Limpopo Thrust Zone. Although this is possible, there is no obvious intensification of the fabric in this area to suggest a localization of strain. Subvertical extrusion of a crustal segment during convergence is a satisfactory account for all the kinematic observations (Fig. 10). A mechanical explanation for this behaviour might be sought in the buttressing effect of the Zimbabwe craton during convergence and the specific crustal rheology of the NMZ, combined with the effect of the anisotropy induced by the gneissic fabric, which dips generally steeply to the south in the vicinity of the Mtilikwe shear zone. A clear result from this study is that the late orogenic evolution of the NMZ did not involve any of the structures that are considered typical of gravitational collapse in modern orogens. The normal faulting on the steeply dipping MSZ is not analogous to normal faulting on, for example, the South Tibetan Detachment Fault, which dips at a shallow angle (Burchfiel et al. 1992). The steep fabrics in the NMZ contrast with the subhorizontal attitude expected for midcrustal collapse features, and particularly with Marshak' s (1999) hypothesis that Archaean orogens might contain belts of subhorizontal fabrics. Horizontal shortening apparently persisted throughout the orogenic evolution not only in this part of the NMZ but elsewhere, as seen in the pervasive steeply dipping fabrics. Thus the shortening noted in the study area is not simply a manifestation of gravitational collapse in the adjacent more internal part of the Limpopo Belt. The lack of typical collapse features in the NMZ may be due to a lack of overthickening by thrusting, as described above. Choukroune et al. (1995) suggested that a lack of thrust overthickening, and the great importance of
Fig. 10. Horizontal shortening, subvertical stretching and extrusion of a crustal segment during late Archaean orogeny in the Northern Marginal Zone, Limpopo Belt. Ornaments as in Figure 3. A-A' marks line of section on Figure 3.
EXTRUSION OF LATE ARCHAEAN GRANULITES magmatic processes in crustal thickening, were characteristic of Archaean orogeny, and this contrast with modem orogenies represented a secular change in orogenic style (cf.. Chardon et al. 1998, 2002). Given the magmatically accreted, juvenile nature of the crust and the particularly high geothermal gradient that must have prevailed in the NMZ during the late Archaean (Kramers et al. 2001), the lack of evidence for gravitational collapse is one of the most distinctive aspects of Limpopo Belt geology. The example of the NMZ indicates that specific boundary conditions allowed for the N - S shortening to be accommodated by vertical extrusion of the hot and ductile crustal section without tectonic overthickening, and adds support to the concept that Archaean orogenesis was, in many aspects, different from modern orogeny.
Conclusions Late Archaean tectonics of the Limpopo Belt in the northern part of the Northern Marginal Zone involved horizontal shortening, vertical extension and subvertical extrusion of crust between gently dipping thrusts and a steeply dipping normal shear zone. The extrusion may have been controlled by the buttressing effect of the Zimbabwe craton and the steeply dipping gneissic fabrics of the NMZ. Normal shear sense is observed on careful scrutiny of steeply dipping fabrics that are parallel to the ubiquitous gneissic structure. Previous research may not have detected such normal sense structures, because they utilize fabric that is conventionally interpreted as due to thrusting with horizontal shortening. The late orogenic fabrics and tectonics of the NMZ are fundamentally different from features that characterize gravitational collapse in the late evolution of modern orogens. The lack of gravitational collapse may have been because the crust was not overthickened by thrusting; the latter may distinguish modem from Archaean orogenesis. We gratefully acknowledge the hospitality of Renco Mine and in particular the underground manager at the time, Shepherd Kadzviti. Thorough reviews by D. Chardon, P. Rey and D. Gapais improved this paper greatly.
References BERGER, M. & ROLLINSON, H. R. 1997. Isotopic and geochemical evidence for extensive intracrustal mixing and homogenization during the Archaean. Geochimicha et Cosmochimica Acta, 61, 48094829.
203
BERGER, M., KRAMERS, J. D. & NAEGLER, TH. F. 1995. Geochemistry and geochronology of charno-enderbites in the NMZ of the Limpopo Belt, Southern Africa, and genetic models. Schweizerische Mineralogische und Petrographische Mitteilungen, 75, 17-42. BIRD, P. 1979. Lateral extrusion of the lower continental crust from under high topography, in the isostatic limit. Journal of Geophysical Research, 96, 10275-10286. BLENKINSOP,T. G., MKWELI, S. etal. 1995. The North Limpopo Thrust Zone (NLTZ): The northern boundary of the Limpopo belt in Zimbabwe and Botswana. Centennial Geocongress, Extended Abstracts, 1, 174-177. BLENKINSOP, T. G. & FREI, R. 1996. Archaean and Proterozoic mineralization and Tectonics at the Renco Mine (NMZ, Limpopo Belt), Zimbabwe. Economic Geology, 91, 1225-1238. BLENKINSOP, T. G., KRONER, A. & CHIWARA, V. 2005. Single stage, late Archaean exhumation of granulites in the Northern Marginal Zone, Limpopo Belt, Zimbabwe, and relevance to gold mineralization at Renco mine. South African Journal of Geology, in press. BURCHF1EL,B. C., CHEN, Z., HODGES, K. V., LIu, Y., ROYDEN, L. H., DENG, C. & Xu, J. 1992. The South Tibetan detachment system, Himalayan orogen: Extension contemporaneous with and parallel to shortening in a collisional mountain belt. Geological Society of America Special Paper, 269, 1-41. CHARDON, D., CHOUKROUNE, P. & JAYANANDA, M. 1998. Sinking of the Dharwar basin (South India): implications for Archaean tectonics. Precambrian Research, 91, 15-39. CHARDON, D., PEUCAT, J.-J., JAYANANDA, M., CHOUKROUNE, P. & FANNING, C. M. 2002. Archaean granite-greenstone tectonics at Kolar (South India): Interplay of diapirism and bulk inhomogeneous contraction during juvenile are accretion. Tectonics, 21, 710-717. CHIWARA, V. 2003. Late Archaean to early Proterozoic tectonics in the northern marginal zone with special emphasis on gold mineralization. Unpublished MSc thesis, University of Zimbabwe. CHOUKROUNE, P., BOUHALLIER,H. & ARNDT, N. T. 1995. Soft lithosphere during periods of Archaean crustal growth or crustal reworking. In: COWARD, M. P. & RIES, A. C. (eds) Early Precambrian Processes. Geological Society, London, Special Publications, 95, 67-86. DEWEY, J. F. 1988. Extensional collapse of orogens. Tectonics, 7, 1123-1139. DE WIT, M. J. 1998. On Archaean granites, greenstones, cratons and tectonics: does the evidence demand a verdict? Precambrian Research, 91, 181-226. DE WIT, M. J., ROERING,C., HART, R. J., ARMSTRONG, R.A., de RONDE, C. E. J. & GREEN, R. W. E. 1992. Formation of an Archean continent. Nature, 357, 553 -562. ENGLAND, P. C. & HOUSEMAN,G. A. 1988. The mechanics of the Tibetan plateau. Transactions of the Royal Society of London, A, 326, 301-320.
204
T.G. BLENKINSOP & A. F. M. KISTERS
ENGLAND, P. C. & HOUSEMAN, G. A. 1989. Extension during continental convergence, with application to the Tibetan plateau. Journal of Geophysical Research, 94, 17561-17579. FREI, R., BLENKINSOP,T. G. & SCHONBERG, R. 1999. Geochronology of the late Archaean Razi & Chilimanzi Suites of granites in Zimbabwe; implications for the late Archaean, tectonics of the Limpopo Belt and Zimbabwe Craton. South African Journal of Geology, 102, 55-63. HAMILTON, W. B. 2004. An alternative Earth. GSA Today, 13, 4-12. HOUSEMAN, G., MCKENZIE, D. P. & MOLNAR, P. 1981. Convective instability of a thickened boundary layer and its relevance for the thermal evolution of continental convergent belts. Journal of Geophysical Research, 86, 6115- 6132. JAMES, P. R. 1975. A deformation study across the
northern margin of the Limpopo Mobile belt, Rhodesia. PhD thesis, University of Leeds. KAMBER, B. S. & BIINO, G. G. 1995. The evolution of high T - l o w P granulites in the NMZ sensu stricto, Limpopo Belt, Zimbabwe - the case for petrography. Schweizerische Mineralogische und Petrographische Mitteilungen, 75,427-454. KAMBER, B. S., BLENKINSOP, T. G., VILLA, 1. M. & DAHL, P. S. 1995. Proterozoic transpressive deformation in the NMZ, Limpopo Belt, Zimbabwe. Journal of Geology, 103, 493-508. KAMBER, B. S., WIJBRANS, J. R., BIlNO, G. G., VILLA, 1. M. & DAVIES, G. 1996, Archaean granulites of the Limpopo Belt, Zimbabwe: one slow exhumation or two rapid events? Tectonics, 15, 14141430. KRAMERS, J. D., KREISSIG, K. & JONES, M. Q. W. 2001. Crustal heat production and style of metamorphism: a comparison between two Archean high grade provinces in the Limpopo Belt, Southern Africa. Precambrian Research, 112, 149-163. KRONER, A. 1981. Precambrian plate tectonics. In: KRONER, A. (ed) Precambrian Plate Tectonics. Elsevier, Amsterdam, 57-90. MARSHAK, S. 1999. Deformation style way back then: thoughts on the contrasts between Archean/Paleoproterozoic and contemporary orogens. Journal of Structural Geology, 21, 1175-1182. MKWELI, S., KAMBER, B. & BERGER, M. 1995. A westward continuation of the Zimbabwe cratonNMZ tectonic break and new age constraints on the timing of the thrusting. Journal of the Geological Society of London, 152, 77-83. NGUURI, T. K. J., GORE, D. E., JAMES, S. J., WEBB, C., WRIGHT, T. G., ZENGENI, O., GWAVAVA, J. A., SNOKE, L. & Kaapvaal Seismic Group. 2001. Crustal structure beneath southern Africa and its implications for the formation and evolution of the Kaapvaal and Zimbabwe cratons. Geophysical Research Letters, 28, 2501-2504.
OBERTHLIR, T., DAVIS, D. W., BLENKINSOP,T. G. & HOHNDORF, A. 2002. Precise U - P b mineral ages, R b - S r and S m - N d systematics for the Great Dyke, Zimbabwe - constraints on crustal evolution and metallogenesis of the Zimbabwe Craton. Precambrian Research, 113, 293-305. ODELL, J. 1975. Explanation of the geological map of the country around Bangala dam. Rhodesian Geological Survey Bulletin, 42, 46. PELLETIER, A., GAPAIS, D., MI~NOT, R.-P. & PEUCAT, J.-J. 2002. Tectonique transpressive en terre Ad~lie au Pal~oprot~rozo'ique (Est Antarctique). Comptes Rendus Geoscience, 334, 505511. PLATT, J. P. & ENGLAND,P. 1994. Convective removal of the lithosphere beneath mountain belts: thermal and mechanical consequences. American Journal of Science, 294, 307-336. REY, P., VANDERHAEGHE, 0. & TEYSS1ER, C. 2001. Gravitational collapse of the continental crust: definition, regimes and modes. Tectonophysics, 342, 435 -449. RIDLEY, J. 1992. On the origins and tectonic significance of the Charnockite suite of the Archaean Limpopo Belt, NMZ, Zimbabwe. Precambrian Research, 55, 407-427. ROERING, C., VAN REENEN, D. D. et al. 1992. Tectonic model for the evolution of the Limpopo Belt. Precambrian Research, 55, 539-552. ROLLINSON, H. R. 1989. Garnet-orthopyroxene thermobarometry of granulites from the notch marginal zone of the Limpopo belt, Zimbabwe. In: DALY, J. S., CLIFF, R. A. & YARDLEY, B. W. D. (eds) Evolution of Metamotphic Belts. Geological Society, London, Special Publications, 43, 331-335. ROLL1NSON, H. R. 1993. A terrane interpretation of the Archaean Limpopo Belt. Geological Magazine, 130, 755-765. ROLLINSON, H. R. & BLENKINSOP, T. G. 1995. The magmatic, metamorphic and tectonic evolution of the Northern Marginal Zone of the Limpopo Belt in Zimbabwe. Journal of the Geological Society of London, 152, 65 -77. SANDIFORD, M. 1989a. Secular trends in the thermal evolution of metamorphic terrains. Earth and Planetary Science Letters, 95, 85-96. SAND1FORD, M. 1989b. Horizontal structures in granulite terrains: a record of mountain building or mountain collapse? Geology, 17, 449-452. STUART, G. W. & ZENGENI, T. G. 1989. Seismic crustal structure of the Limpopo mobile belt, Zimbabwe. Tectonophysics, 144, 323-335. T1KOFF, B. & TEYSSIER, C. 1994. Strain modelling of displacement-field partitioning in transpressional orogens. Journal of Structural Geology, 16, 1575-1588.
Synergistic effects of melting and deformation: an example from the Variscan belt, western France MICHAEL BROWN
Laboratory for Crustal Petrology, Department of Geology, UniversiO' of Maryland, College Park, MD 20742-421 l, U.S.A. (e-mail: mbrown @geol. umd. edu) Abstract" In the Vannes and St. Nazaire regions of the central part of the Domaine Sud-Armoricain, Variscan belt, western France, the lowermost tectonic unit is exposed as structural culminations composed of supracrustal migmatites that were deformed and metamorphosed along a multistep clockwise P-T path during Carboniferous time; peak P-T was around 800 ~ at 9 kbar. The strain field that emerged under subsolidus conditions during prograde metamorphism controlled the initial distribution of granite melt produced by suprasolidus mica breakdown; the limited retrograde reaction of peritectic garnet indicates that melt loss occurred around the metamorphic peak. A second episode of melt production occurred during the retrograde evolution due to a decompression event that led to interconnection of melt in a mesoscale network of deformation bands and formation of ductile opening-mode fractures, as evidenced by layer-parallel and transverse leucosomes linked with petrographic continuity to granite in dykes. The preservation of peritectic cordierite with only limited associated leucosome and the occurrence of pucker structures without leucosome both indicate that melt loss occurred during the second event. Dykes vary from centimetric (common) to hundreds of metres in width (rare), and exhibit scale-invariance over a limited range of measurements; larger dykes are inferred to have fed upper crustal plutons. Melt extraction may have been a self-organized critical phenomenon, but this remains to be demonstrated satisfactorily in nature. Fugitive melt was trapped in the vicinity of the brittle-ductile transition zone and emplaced laterally along horizons reactivated as extensional detachments. A feedback relation is postulated between dextral transtensive deformation, decompression melting and lower crustal doming, and between dome amplification, melt extraction and emplacement in developing extensional detachments and core complex formation.
The infrastructure of orogens is commonly exposed as a series of core complexes; typically these are composed of high-grade metamorphic rocks - migmatites and granulites - that record evidence of melting or have residual compositions. In contrast, late orogenic granites that include a crustal component are a common feature of the suprastructure of orogens (Brown 2001a, b). Many studies have documented an intimate relationship between presence of melt and localization of deformation, and the feedback relations between melting and deformation (e.g. Brown & Rushmer 1997; Brown & Solar 1998a). Further, rock deformation experiments have shown that the presence of melt weakens rock materials (e.g. Van der Molen & Paterson 1979; Rutter 1997; Gleason et al. 1999; Holyoke & Rushmer 2002), so that a synergistic relationship between melting and/or magmatic intrusion and deformation is to be expected
(e.g. Parsons & Thompson 1993; Hill et al. 1995; McKenzie & Jackson 2002), and it has been argued that melt plays an important role in the uplift and exhumation of orogenic belts (e.g. Hoilister 1993; Brown & Dalhneyer 1996; Vanderhaeghe & Teyssier 2001; Teyssier & Whitney 2002). For these reasons, the dynamic implications of crustal anatexis and melt extraction and emplacement need to be better understood. In anatectic migmatites and residual granulites, interconnection of leucosome may be demonstrated from grain- to vein-scale (Marchildon & Brown 2001, 2002; Sawyer 2001; Guernina & Sawyer 2003); this leucosome is inferred to record remnants of the melt flow network (Brown 1994, 2001a, b; Oliver & Barr 1997; Brown et al. 1999; Sawyer 2001; Guernina & Sawyer 2003; Marchildon & Brown 2003). However, the connection between melt flow networks and magma ascent conduits remains
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics: from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 205-226. 0305-8719/05/$15.00 ~(~ The Geological Society of London 2005.
206
M. BROWN
enigmatic, although theoretical and phenomenological models have been proposed (Petford & Koenders 1998; Bons et al. 2001, 2004) and ductile fracture has been postulated (Brown 2004). Orogens are complex systems, in which it is necessary to evaluate processes holistically at various scales; complex systems evolve toward an optimal dynamic structure to dissipate energy. This is expressed at the scale of the orogen by the internal self-organization of tectonic elements and the thermal evolution (Hodges 2000) and may be expressed at the scale of the outcrop in lower crustal rocks by the patterning of structures and leucosome networks in lower crustal migmatites and granulites (Brown & Solar 1999; Guernina & Sawyer 2003; Marchildon & Brown 2003). Furthermore, core complexes within orogens generally record evidence of a positive feedback relation between decompression and crustal anatexis, which suggests that the vertical transfer of lower crust during core complex formation is an efficient mechanism for advection of heat and cooling of orogens (Teyssier & Whitney 2002). The focus of this paper is on melting in continental crust, melt extraction and ascent, and the role of fugitive melt in exhumation of the lower crust of orogens. The synergistic effects of melting and deformation are illustrated using evidence from lower crustal migmatites and leucogranite associated with Upper Carboniferous core complex formation in the Domaine SudArmoricain in the Variscan belt of western France (Gapais et al. 1993; Brown & Dallmeyer 1996; Marchildon & Brown 2003).
The Variscan belt of western France The evolution of the European Variscides was complex - a consequence of interactions between Laurasia and Gondwana (Dewey & Burke 1973; Matte 1986, 1991; Rey et al. 1997). In western France, the Variscan cycle is interpreted to have comprised Ordovician rifting to form the Galicia-Southern Brittany Ocean, Silurian subduction and consequent ocean closure in the Devonian, and Carboniferous collision between a promontory of Gondwana and Armorica that involved closeto-orthogonal sinistral transpressive deformation (Brun & Burg 1982; Pin & Peucat 1986; Burg et al. 1987; Audren 1990; Jones 1991; Dias & Ribeiro 1995; Matte 2001). Subsequent Upper Carboniferous intracontinental displacement involved highly oblique dextral transtensive deformation and exhumation of lower crustal rocks associated with the emplacement of
leucogranites (e.g. Faure & Pons 1991; Gapais et al. 1993; Geoffroy 1993; Burg et al. 1994;
Brown & Dallmeyer 1996).
Geology of the Domaine Sud-Armoricain The rocks discussed here are part of the Domaine Sud-Armoricain (DSA; Fig. 1), a NW-SE-trending belt of supracrustal rocks deformed and metamorphosed during the Variscan orogeny (Vidal 1973; Autran & Cogn6 1980; Brun & Burg 1982; Peucat 1983; Gapais et al. 1993; Brown & Dallmeyer 1996). The DSA is separated from the Domaine Centre-Armoricain, Domaine Ligerien and Domaine Nantais by the southern branch of the South Armorican Shear Zone (SASZ; Fig. 1); together these domains comprise the Variscan belt of western France (Truffert et al. 2001). The four domains most probably represent tectonostratigraphic terranes juxtaposed during the Devonian, although there are different views concerning the tectonic evolution, the significance of thrusting versus normal-sense displacements and sinistral versus dextral displacements, and the amount of displacement between terrane elements (e.g. Brun & Burg 1982; Vauchez et al. 1987; Gapais et al. 1993; Faure et al. 1997; Shelley & Bossi6re 2000; Cartier et al. 2002). T e c t o n o - s t r a t i g r a p h i c units in the D S A
In the DSA, the protracted Palaeozoic tectonic evolution ultimately resulted in extensive deformation, metamorphism and partial melting of Neoproterozoic supracrustal rocks that are now exposed as a core of migmatitic gneiss domes. In the central part of the DSA, around Vannes and St. Nazaire (Fig. 1), these migmatites preserve a mineralogical record of metamorphism and melting along an overall clockwise P - T path (peak P - T of 9 kbar, 800 ~ Brown 1983; Jones 1988; Jones & Brown 1989, 1990), with a stepped retrograde P - T segment involving decompression and a second episode of melt generation (Brown & Dallmeyer 1996). Figure 2 is a P - T pseudosection constructed for a postulated residual composition (i.e. after some melt loss around the metamorphic peak) to illustrate the retrograde evolution relevant to the late orogenic deformation discussed in this paper; evidence for the full P - T evolution is discussed in detail in Johnson & Brown (2004). This high-grade metamorphic core is bounded to the north by the NW-SE-trending transcurrent SASZ (J6gouzo 1980), and is separated from overlying units composed of lower-grade rocks to the southwest by shallow-dipping detachments
SYNERGY BETWEEN MELTING AND DEFORMATION
207
I
50 km
I
2ow
48~+ DOMAINE SUD-ARMORICAIN
- ~
Q-T Tertiary& QuaternaryRocks
G ~ W !'--
MesozoicSedimentaryRocks
~ , ~ ~ .
V .. M \'>..~
~
Blueschists r~
,
Qui
Belle-IleGroup
~
i
\
OOMAINECENTRE"4RMORICAIN F
•~
DOMAINE
v
N-E
~
F'~ VilaineGroup
LIGURIAN
'~
"~.->"
Variscangranites
DOMAINE NANTAIS
47oN~.W I-
Anatecticrocks Eciogites(Essartscomplex)
~-.~
UndifferentiatedPaleozoicRocks QSZ
SASZ [
\ St.G
LSO i 15 km i
'"~,,
m
Fig. 1. Geological map of the Domaine Sud-Armoricain, a NW-SE-trending belt of supracrustal rocks deformed and metamorphosed during the Paleozoic Variscan orogeny. The southern branch of the South Armorican Shear Zone (SASZ) separates the Domaine Sud-Armoricain from the three domains to the northwest (unomamented with domain names); together these four domains comprise the Variscan belt of western France. Granites: PA, Pony Abb& P1, Ploemur; Qui, Quiberon; Que, Questembert; S, Sarzeau; Gu, Gu&ande. Place names: L, Lorient; G, Ile de Groix; BI, Belle Ile; V, Vannes; M, Golfe du Morbihan; St.N, St. Nazaire: BC, Bois de C6n6; St.G, St. Gildas; LSO, Les Sables d'Olonne. QSZ, Quiberon Shear Zone. The W - E sketch section is modified after Gapais et al. (1993).
and associated crustally derived granites (Gapais et al. 1993). Amphibolites of the adjacent Essarts
Complex contain relics of eclogite facies rocks (15-20kbar, 600-800~ Godard 1988); the Essarts Complex is separated from the highgrade metamorphic core by tectonic contacts (Godard 2001 ). The structurally overlying Vilaine Group is dominated by apparently unmigmatized interbedded metaclastic and meta-tuffaceous rocks with amphibolite horizons; it records uppermost amphibolite facies metamorphism around the Vilaine estuary (peak P - T of 7-9 kbar, 650 :C; Triboulet & Audren 1985, 1988), but in Vend6 the metamorphic grade decreases upward through the Group (e.g. Bossibre 1988; Goujou 1992). The lower part of the overlying Belle-Ile Group is characterized by highly strained, low-grade siliceous metavolcanic and
metavolcaniclastic rocks (the Vend6e porphyroids) and metapelites (e.g. Audren & Plaine 1986; Le H6bel et al. 2002a; Schultz et al. 2002); these rocks record moderate-P-low-T metamorphism (peak P - T of 8kbar, 350400 "*C; Le Hdbel et al. 2002b). Inferred to overlie these units are the blueschist rocks and eclogites of oceanic affinity exposed on the Ile de Groix and in the Bois de Cen6 (e.g. Quinquis & Choukroune 1981; Bernard-Griffiths et al. 1986; Shelley & Bossibre 1999; Schultz et al. 2001); these rocks record high-P-low-T metamorphism (Upper Unit peak P - T of 1618 kbar, 450-500 "C; Lower Unit peak P - T of 14-16 kbar, 400-450 ~ Triboulet 1974, 1991; Guiraud et al. 1987; Djro et al. 1989; Bosse et al. 2002; Ballbvre et al. 2003). Contacts between units are flat-lying. In the central part of the DSA, the Belle-Ile Group
208
M. BROWN
Fig. 2. MnNCKFMASHP-T pseudosection from Johnson and Brown (2004) constructed for a residual bulk composition that assumes melt loss of mol.% of the melt produced at the metamorphic peak (see Johnson & Brown 2004 for further details). This diagram is only appropriate for consideration of the retrograde phase equilibria. The principal effect of melt (and H20) loss is the elevation of solidus temperatures and an increased stability of garnet. Two possible decompression P-T segments consistent with petrological and thermobarometric constraints summarized by Johnson & Brown (2004) are shown in black. Dashed lines are tool.% melt isopleths for this residual bulk composition; the darkest fields are quinivariant (F = 5).
generally is separated from the underlying highgrade rocks by subhorizontal tabular plutons of two-mica leucogranite. S-C mylonitic fabrics defined by a flat-lying schistosity, a single set of dominant shear bands and a strong mineral elongation lineation are common within these teucogranites, particularly in their lower parts (e.g. Bouchez et al. 1981; Audren 1987; Gapais et al. 1993). These features are interpreted to indicate that granite emplacement was synkinematic (cf. Gapais 1989), and S-C fabrics indicate normal-sense displacement. Following Gapais et al. (1993), the first-order normal-sense feature that is approximately parallel to the coastline is termed the Quiberon Shear Zone (QSZ; Fig. 1). The overall architecture defines windows of high-T-low-P rocks that represent structural culminations surrounded by the highP - l o w - T units in structural depressions (Fig. 1). Age relations
Recent geochronology on the porphyroids and blueschists of the Belle-Ile Group indicates that
the peak of high-P-low-T metamorphism was at 370-360 Ma, with exhumation and cooling c. 350 Ma (e.g. Bosse et al. 2000; Le H~bel 2002). In contrast, geochronological data from the high-T-moderate-to-low-P migmatites and anatectic granites of the SBMB suggest a rapid evolution in the Upper Carboniferous, since age data on monazite (U-Pb), hornblende (4~ 39Ar), mica ( R b - S r and 4~ and apatite (fission tracks) all lie in the range c. 320290 Ma (e.g. Vidal 1973; Carpena et al. 1979; Peucat 1983; Brown & Dallmeyer 1996). Similar 4~ ages of c. 300 Ma on hornblende in amphibolites from the SBMB and the Vilaine Group from the Vilaine estuary (Brown & Dallmeyer 1996) suggest these two units have a common retrograde evolution, hnprecise U - P b ages determined on metamorphic and anatectic zircon have been used to argue for an earlier age of c. 420-380 Ma for the peak of high-T metamorphism (Peucat 1983; Brown & Dallmeyer 1996), but exactly what these data record is unclear in relation to information that suggests an important Upper Carboniferous evolution for these rocks, and the greater likelihood of inheritance in zircon in comparison with monazite (e.g. Rubatto et al. 2001). One implication of these data is that the supracrustal rocks of the lower crust may have contained melt for some tens of millions of years (cf. Gerdes et al. 2000). Ages of two-mica granites emplaced synkinematically along major strike slip and extensional shear zones in South Armorica help constrain the timing of displacement along these shear zones. Bernard-Griffiths et al. (1985) documented an apparent younging of R b - S r whole-rock isochron ages of these granites from north to south across the SASZ, from c. 345 Ma to c. 300 Ma. Older ages are from granites exposed north of (c. 3 4 5 - 3 4 0 M a ) and along (c. 330-320 Ma) the SASZ; they are interpreted to record the time of crystallization and likely record the age of emplacement. Therefore, the 3 3 0 - 3 2 0 M a ages may be used to constrain timing of the initiation of the dextral strike-slip displacement along the SASZ, since these granites exhibit subhorizontal mineral elongation lineations and dextral S-C fabrics (Berth~ et al. 1979); 4~ analysis on mica from granites along the SASZ yields ages of 310-300 Ma, which are interpreted to date recrystallization during ongoing subsolidus deformation (unpublished results of Ruffet 2001, quoted in Le H~bel 2002). Two-mica granites exposed south of the SASZ occur as shallow tabular plutons along contacts between tectono-stratigraphic units; these yield R b - S r ages of c. 3 0 5 - 3 0 0 M a ,
SYNERGY BETWEEN MELTING AND DEFORMATION consistent with 4~ ages on muscovite (Brown & Dallmeyer 1996), which serve to constrain timing of the initiation of extensional displacement based on S-C fabrics in the granites (Gapais et al. 1993), and suggest rapid exhumation and cooling.
Summary
The tectono-metamorphic evolution of the DSA is composed of two principal stages (cf. Brown & O'Brien 1997): an Upper Devonian to Lower Carboniferous high-P-low-T stage that involved thrusting and piling up of nappes during a subduction to continental collision transition; and an Upper Carboniferous high-T metamorphism and associated granite magmatism related to regional sinistral transpressive and dextral transtensive deformation and core complex formation. It has been speculated that postcollisional slab detachment occurred to motivate exhumation of the lower crustal units and core complex formation in the DSA (Brown & Dallmeyer 1996). However, interpretation of newly acquired geophysical data has led to the suggestion that a relict slab remains beneath Central Armorica (Judenherc et al. 2002, 2003; Bitri et al. 2003; Gumiaux et al. 2004). If the
209
slab did not break off or if detachment did not lead to the slab sinking into the mantle to allow asthenospheric heat to reach the base of the thinned lithosphere under the over-riding plate (van de Zedde & Wortel 2001), then peak metamorphic conditions may reflect thermal relaxation of a perturbed temperature profile enhanced by radiogenic heating, and exhumation must have been motivated by some mechanism other than slab detachment.
Anatectic rocks of the Domaine Sud-Armoricain Anatectic rocks of concern in this paper are exposed around Vannes in the Golfe du Morbihan, along the coast of the Presqu'~le de Rhuys and along the coast WSW of St. Nazaire (Fig. 1). These rocks are migmatites that comprise predominantly metatexite (Fig. 3) with minor diatexite (Brown 1983; Audren 1987; Jones & Brown 1990); they are traversed by centimetric to metric dykes of granite (Fig. 4), and several 'mega-dykes' up to 1 km wide. The petrological and structural characteristics of these rocks have been described in detail by a number of authors (Cogn6 1960; Brown 1983; Audren 1987; Jones 1988; Jones & Brown 1989, 1990;
Fig. 3. Features associated with melting and inferred melt-bearing structures in migmatites from Petit Mont, Morbihan. (a) Stromatic migmatite showing foliation boudinage; (b) stromatic migmatite showing compaction (C), dilation (D) and shear (S) bands infilled with leucosome; (c) steeply oriented surface parallel to So-Sl (note subhorizontal elongation lineation) to show vertical extent of leucosome in interboudin partitions.
210
M. BROWN
Fig. 4. Features associated with melt extraction in migmatites from Petit Mont, Morbihan. (a) View roughly NW of 0.5 m-wide dykes of granite in stromatic migmatite; (b) dykes of granite frozen close to site of formation in stromatic migmatite; (e) subhorizontal surface through stromatic migmatite to show synchronous centimetric granite dykes that intersect in a common steeply inclined 'pipe-like' channel structure (with Brunton compass for scale); there is only limited evidence for cross-cutting relationships, suggesting that material now in the dykes was molten at the same time.
Audren & Triboulet 1993; Brown & Dallmeyer 1996; Marchildon & Brown 2003; Johnson & Brown 2004). The migmatitic lower crustal unit followed a multistep clockwise P-T path (Brown 1983; Jones & Brown 1990; Audren & Triboulet 1993). This involved a two-stage prograde evolution to the metamorphic peak at 9 kbar, 800 C (Johnson & Brown 2004) with production of about 25 vol.% melt via low-aH20 muscoviteand biotite-consuming reactions producing
peritectic garnet (Fig. 5a). Johnson & Brown (2004) estimate that approximately 60% of the melt produced at the metamorphic peak was extracted and lost from the system, evidence of which is the variably incomplete retrogression of peritectic garnet associated with minimal leucosome (Fig. 5a). This was followed by a stepped retrograde evolution (Fig. 2) that began with erosion controlled conductive cooling, recorded by the partial reaction of garnet with melt to form biotite and sillimanite, was interrupted by
SYNERGY BETWEEN MELTING AND DEFORMATION
211
Fig. 5. Petrographic features in stromatic migmatite that support melt loss, Presqu'i'le de Rhuys, Morbihan. (a) Virtually pristine (unretrogressed) peritectic garnet in leucosome-poor stromatic migmatite, Petit Mont; (b) stromatic migmatite, Port Navalo Plage, note partially reacted garnet in foliation (lower right-hand side) and peritectic cordierite in interboudin partition (centre left) with minor leucosome; (c) stromatic migmatite, Port Navalo Plage, note peritectic cordierite in interboudin areas and adjacent to shear bands mostly without any associated leucosome; (d) stromatic migmatite, Port Navalo Plage, note cordierite in interboudin partition without much leucosome; (e) pucker structure in which layering is 'sucked' into interboudin partitions reflecting melt loss from these sites, note thickening of leucosome stromata in boudin neck, Petit Mont; (f) folded stromatic migmatite, Port Navalo Plage, thicker leucosome contains peritectic cordierite (throughout but particularly at bottom left-hand side) to suggest that decompression melting in which biotite breaks down to produce peritectic cordierite with melt was synchronous with F3 folding.
212
M. BROWN
a steep decompression segment that involved a second episode of melt production evidenced by the presence of centimetric peritectic cordierite associated with minimal leucosome, located in dilational sites but recording melt loss (cf. Bons 1999), and was terminated by rapid cooling below the solidus (Brown & Dallmeyer 1996). The metamorphic evolution may have taken on the order of 30 m.y. or more, with much of this evolution being suprasolidus; this is consistent with residence times for melt in anatectic crust of other studies (e.g. Rubatto et al. 2001 ). Structural evolution
Early progressive sinistral transpressive deformation in the migmatites is recorded by two sets of superimposed folds, inferred to have produced flat-lying structures in the lower crust prior to doming, and later dextral transtensive deformation by folding that defines the regional map pattern of foliation form lines (cf. Brun & Burg 1982; Audren 1987; Matte 2001; Cartier et al. 2002). Isoclinal intrafolial folds (F1) of compositional layering (So) have an associated axial planar foliation (S~); the resulting composite So-S1 foliation is typically parallel to compositional layering. Folding of So-S~ produced a generation of roughly N W - SE-trending regional synformal and antiformal F2 fold structures (Cogn6 1960; Audren 1987), but a new penetrative axial planar foliation generally is not developed. F 2 folds are close to tight, with subhorizontal hinge lines and subvertical axial surfaces, which accounts for the variable dip of the So-SI composite foliation. During the early part of the late progressive dextral transtensive deformation, which also involved doming, apparent extension related to decompression melting has resulted in widespread foliation boudinage (Fig. 3a) and formation of ductile dilation and shear bands (Fig. 3b) in multilayers in migmatite, and accumulation of melt in and loss of melt from the resulting interboudin partitions and shear surfaces, evidenced by more or less discontinuous cross-cutting leucosomes in which centimetric peritectic cordierite is locally developed (Fig. 5b). The absence of significant retrogression of the cordierite, together with the limited volume of associated leucosome (Fig. 5b), which is much less than the volume of melt predicted by mass balance, indicates loss of a substantial proportion of the melt accumulated at these sites (e.g. Johnson & Brown 2004). In some cases, cordierite in host migmatite is isolated from leucosome because all melt
has migrated away from these sites (Fig. 5c, d); this conclusion is further supported by pucker structures without leucosome and thickening of leucosomes in interboudin necks (Fig. 5e). The latter part of the third deformation produced open to tight, generally asymmetric F3 folds that have steep hinge lines and subvertical axial surfaces. Regional- and decametre-scale F3 folds account for the scatter in strike of the So-Sl foliation (Cogn6 1960; Audren 1987). Centimetre-scale F3 folds are developed only locally, primarily in metatexites, and preferentially in the short limbs of metric or decametric F3 folds or in discrete zones with a dextral sense of shear; an axial planar foliation, $3, is commonly observed where F 3 folds are centimetric. F3 folding was likely coeval with dextral displacement along the SASZ during the later stages of dextral transtension and the progressive localization of strain; in zones of high late-D3 strain, F3 folds have become disrupted and isolated as intrafolial relicts. All three phases of deformation fold migmatitic layering (cf. Jones 1988, 1991). A subhorizontal or shallow-pitching quartz-feldspar aggregate lineation, inferred to record a principal finite stretch, is ubiquitous throughout the area; it is particularly well defined on foliation-parallel surfaces. In sections parallel to stromatic layering and lineation, interboudin leucosomes are perpendicular to the lineation and continuous over tens of centimetres (Fig. 3c), consistent with boudinage and the development of meltbearing interboudin partitions being synchronous with the development or the modification/reactivation of the lineation. The trend of this lineation varies with the orientation of the dominant foliation (So-S1) on which it is observed; it is folded by the F3 folds together with the stromatic leucosomes. Cordierite is present in some of the folded leucosomes (Fig. 50, which suggests that F3 folding occurred during decompression or subsequent cooling, but under suprasolidus conditions that allowed melt migration from fold limbs to hinges and melt loss along the steeply inclined hinge lines (cf. the 'passive migration' mechanism of Barraud et al. 2004). The F3 folding indicates that by late in the D 3 deformation the strike of the migmatitic layering was oriented in the shortening sector of the D3 finite strain ellipsoid. However, early D3 layerparallel extension and decompression melting, as reflected in boudinage of compositional layering and foliation, is inferred to have occurred when the migmatitic layering was oriented in the lengthening sector of the D3 infinitesimal strain ellipsoid. Rotation of the migmatitic layering both counterclockwise and back to
SYNERGY BETWEEN MELTING AND DEFORMATION approximately horizontal about a N W - S E axis would bring the perpendicular to the interboudin partitions into the lengthening sector of the infinitesimal strain ellipsoid during the early stages of dextral transtension. In this orientation the lineation may represent an early D3 feature, or possibly a D 1 - D 2 feature modified during D3 deformation. Thus, the apparent contradiction of layer-parallel extension followed by layerparallel shortening and folding of the lineation is a direct consequence of doming and core complex formation during progressive dextral transtensive deformation, which rotated the layering clockwise and from shallowly dipping to steeply dipping. During this geometrically complex progressive deformation the layering is inferred to have migrated through the lengthening sector of the infinite and finite strain ellipsoids and across the surfaces of no infinitesimal and no finite longitudinal strain into the shortening sector of the finite strain ellipsoid. On the outcrop, leucosomes generally do not show evidence of solid-state deformation. In thin section, magmatic microstructures, such as rational faces of feldspar, and quartz or feldspar films inferred to pseudomorph intergranular melt are common in leucosomes, suggesting that much of the penetrative deformation that affected these rocks either pre-dated final melt crystallization, or was partitioned away from mostly crystalline leucosomes (Brown & Dallmeyer 1996; Marchildon & Brown 2003). Thus, migmatites of the Port Navalo area appear to have hosted melt continuously from prior to or during F1 folding until F3 folding. F o r m e r m e l t - b e a r i n g structures
Leucosome- and granite-filled structures record the melt flow network; these structures include ram- to cm-scale foliation-parallel and foliation-discordant leucosome stromata (reflecting compaction-enhanced layering, interboudin partitions and extensional shear surfaces) and cm- to m-scale dykes of granite (representing inferred ascent conduits). Petrographic continuity (mineralogy, mode and microstructure) of the infilling material from structure to structure implies that some proportion of this material crystallized from a continuous melt-beating network (Figs 3 and 6). Brown (2004) has suggested that leucosome networks in migmatites and residual granulites are analogous to deformation band networks composed of shear bands (Antonellini et al. 1994), compaction bands (Mollema & Antonellini 1996) and dilation bands (Du Bernard et al. 2002); they
213
represent a melt accumulation network (Figs 3b and 7a). Using quantitative data on the one- and two-dimensional distribution of inferred meltbearing structures, Marchildon & Brown (2003) investigated the transitions from fabricparallel to network to conduit flow in an attempt to understand the evolution of the architecture (connectivity and perrneability) of meltbearing systems. They obtained information about thickness and spacing distributions of layer-parallel leucosome from one-dimensional line traverses across layering on flat surface exposures of stromatic migmatite. Layer-parallel leucosome thicknesses mostly fall in the range 1-10 mm, with an upper limit around 2 0 30 mm. The number of thicker veins decreases abruptly with increasing thickness, which is inconsistent with scale-invariance; spacing distributions also are not scale-invariant. The lack of clustering of stromata in the data of Marchildon & Brown (2003) is inconsistent with an origin by fracturing, since fractures tend to be clustered or have regularly spaced distribution around a single large fracture. An origin by mass transfer down gradients in melt pressure generated by heterogeneous deformation of an anisotropic protolith is consistent with the scale-dependant nature of the data and is preferred (cf. Brown et al. 1995). Qualitative observation of inferred meltbearing structures in mutually perpendicular two-dimensional flat surface exposures from the same outcrop reveals anisotropy of the leucosome network related to the subhorizontal stretching lineation. On lineation parallel surfaces, traces of stromatic leucosomes are continuous and the leucosomes have smooth essentially straight edges (Fig. 7a), whereas on lineation normal surfaces, traces of stromatic leucosomes are more complex, with less continuity and more irregular outlines (Fig. 7b). Overall, the three-dimensional anisotropy suggests an apparent constrictional leucosome structure with the long axis parallel to the lineation (Fig. 7c). Marchildon & Brown (2003) report results of semi-quantitative analysis of the outlines of these leucosomes in two dimensions using the box counting method, which corroborate the inferred anisotropy, and show that leucosome morphology exhibits only limited scale invariance. This result is not surprising given the strong control exerted by the anisotropy on leucosome network morphology. The anisotropic nature of the stromatic leucosomes suggests that permeability in the plane of the foliation was larger parallel to lineation than perpendicular to lineation, which further
214
M. BROWN
Fig. 6. Petrographic continuity of leucosome and granite, Petit Mont, Morbihan. (a) Dyke of granite in stromatic migmatite; (b) detail of (a), located immediately below lens cap, to show connectivity of leucosome in stromatic migmatite with granite in dyke (e.g. left of lens cap); (e) granite dyke in stromatic migmatite; (d) close-up of (c), to the left of the pen tip, to show petrographic continuity between leucosome in migmatite and granite in dyke.
supports the synchroneity of deformation and leucosome topology. It may be inferred that melt flowed in the plane of the lbliation parallel to the lineation to interboudin partitions in response to gradients in melt pressure or that
diffusive mass transfer through interconnected melt enabled melt accumulation in interboudin partitions (cf. Hand & Dirks 1992). However, the growth of peritectic phases in deformation bands, particularly interboudin partitions
SYNERGY BETWEEN MELTING AND DEFORMATION
215
Fig. 7. Binary maps of inferred melt-bearing structures in (a) subhorizontal (D, dilation band; S, shear band; and C, compaction band) and (b) subvertical exposures, and (c) to show the three-dimensional relationship and apparent constrictional topology of the leucosome, Petit Mont, Morbihan (from Marchildon, N. and Brown, M. 2003. Spatial distribution of melt-bearing structures in anatectic rocks from southern Brittany, France: implications for melt-transfer at gram-scale to orogen-scale. Tectonophysics,364, 215-235. Copyright 2003 Elsevier B.V., with permission).
(Fig. 5), connotes an important contribution by diffusive mass transfer in the local concentration of melt.
Granite dykes Centimetric to metric granite dykes are abundant at outcrop; they commonly cross-cut structures in the migmatites (Figs 4 and 6). The volumetric importance of the dykes varies in space, but they may represent up to 20% by area of an outcrop. Individual dykes are planar and typically unaffected by subsequent deformation,
except locally, indicating that emplacement post-dated the episodes of folding described above. At map scale, 'mega-dykes' of several hundreds of metres width may represent the roots of eroded subhorizontal tabular granites (e.g. Port Navalo and lie aux Moines). Granite in these dykes varies from medium- to coarse-grained. It comprises quartz, which is locally intersertal, oscillatorily zoned plagioclase (An 15- 20), K-feldspar (microcline or perthite), biotite and muscovite. In addition, cordierite, garnet and andalusite may be present, and apatite, zircon and sometimes monazite occur
216
M. BROWN
as accessory phases. The microstructure is granitic, equigranular to porphyritic, with euhedral to subhedral feldspar and anhedral quartz. Within individual dykes, spatial variation in grain size is limited and finer-grained margins generally are not observed, suggesting approximate thermal equilibrium between melt in the fracture and host melt-bearing crust. The nature of the contacts between granite in the dykes and leucosome in the host migmatites is particularly instructive. On the outcrop, the granite dykes generally show discordant contacts with host migmatites, and individual dykes intersect one another (Fig. 4c). However, the modal mineralogy and microstructure of leucosome located in layer-parallel and cross-cutting structures are indistinguishable from the modal mineralogy and microstructure of granite in dykes (Fig. 6). This petrographic continuity is interpreted to mean that these structures once hosted a continuous melt-beating network and suggests that the material in leucosomes and in the dykes underwent final crystallization at the same time. This inference does not mean that leucosomes (or necessarily granite in dykes) have liquid compositions; it is clear that leucosomes comprise residual minerals (quartz and feldspar cores), cumulate material (feldspar rims) and material crystallized from derived liquid, and that dykes record minimum conduit widths compared with the likely aperture width at peak flow. The lack of evidence of dislocation creep, demonstrated by the preservation of magmatic microstructures and the limited plastic deformation of quartz, indicates that crystallization of melt was not followed by significant penetrative deformation. These observations are interpreted to indicate that a network of connected meltbearing structures, represented by layer-parallel and discordant leucosomes and granite dykes, existed in these rocks late in the deformation history. As temperature declined below the solidus, subsequent deformation was partitioned away from the core of the DSA and appears to have been localized into the weaker Upper Carboniferous leucogranites (Hanmer 1997), where significant subsolidus strain is recorded by mylonitic fabrics (Jtgouzo 1980; Jtgouzo & Rossello 1988; Gapais et al. 1993) and is reflected in 4~ mica ages. The petrographic continuity between leucosome in the deformation band networks and granite in dykes suggests that the ascent conduits represent either structures analogous to largescale dilation bands formed by opening-mode failure along zones of localized porosity increase or ductile opening-mode fractures formed by pore growth and coalescence of melt pockets
(e.g. Regenauer-Lieb 1999; Eichhubl et al. 2001; Du Bernard et al. 2002; Eichhubl & Aydin 2003). The dykes fill macroscopic fracture-like discontinuities with a preferred orientation independent of anisotropy, which suggests that dyke orientation is controlled by stress; this feature characterizes the process of dyke formation as a fracture phenomenon. Additional characteristic features of the dykes include the nature of the tips, which are blunter than expected for brittle fracture, the open zigzag geometry close to the tips (Fig. 4b), and the petrographic continuity between leucosome in host and granite in dykes. In particular, the zigzag propagation paths at the tips of dykes, which may originate from transverse structures such as interboudin partitions and shear surfaces, point to ductile fracture as the mode of formation of these conduits (Brown 2004). The thickness of dykes > 10 cm wide shows a power law distribution with an exponent of 1.11 (Fig. 8), comparable to values of 1.1 reported by Bons et al. (2004; who also reported an exponent of 1.9 for one suite of veins). This width-distribution relationship suggests the dykes may be scale-invariant, although the dataset is small (87 dykes) and the range of observations is only two orders of magnitude (cf. Bons et al. 2004). The largest dykes measured along the traverse were 3 m and 5.5 m wide, within the range of critical dyke widths for flowing magma to
1000
exponent = 1.11
r ii Q
z E
10
,,,,i
~.
J
1
10
100
Thickness (cm) Fig. 8. Cumulative frequency plots of dyke thickness (n = 87) from a traverse along the coastline around Port Navalo Plage and le Petit Mont, Morbihan. The number of dykes wider than a particular value is plotted against that value, and a best fit line through the data above 10 cm wide defines a power law relationship with an exponent of 1.11.
SYNERGY BETWEEN MELTING AND DEFORMATION advect heat faster than conduction through the walls and avoid freezing close to the source (Clemens 1998). Dykes of this size each may transport > 1 0 0 0 k m 3 of magma in c. 1200 years, assuming a horizontal length of 1 km (Clemens 1998). This fugitive melt is inferred to have fed crustally-derived plutons at higher crustal levels, such as the Ploemeur, Quiberon, Sarzeau, and Gu~rande granites.
Structural analysis of former melt-bearing structures and dykes Figure 9 shows the orientation of the primary metamorphic fabric and migmatite layering (So-S1), F3 folds, cumulative plots of the strike orientation of centimetric leucosome-filled shear bands and the orientation of centimetric
217
to metric granite dykes. The distribution and orientation of small-scale shear surfaces is approximately congruent with layer-parallel extension as recorded by petrographically continuous leucosome-filled interboudin partitions, assuming rotation of the migmatitic layering both counterclockwise and back to approximately horizontal about a N W - S E axis (as described earlier). In the DSA migmatites the dykes show a range of orientations; most trend approximately NNW and dip steeply to moderately to the W, defining a point maximum (Fig. 9). Audren (1987) interpreted similar results as reflecting the synchroneity of dyke formation with dextral strike-slip deformation along the SASZ, the NNW trend being at low angle to the short axis (i.e. minimum principal finite stretch) of the regional
Fig. 9. The orientation of structural elements related to Carboniferous deformation measured in outcrops around the Presqu'~le de Rhuys, Morbihan (equal area lower hemisphere projection).
218
M. BROWN
finite strain ellipsoid associated with this deformation. The geometrical relationship between the overall strike of granite dykes and the orientation of the SASZ is shown in Figure 10. It has been argued that in the transition from grain-scale melt flow to channelized flow in veins the direction of melt flow will be controlled by elements of the metamorphic fabric, particularly the foliation and lineation (Brown & Solar 1998a; Brown et aL 1999). In zones of highly oblique transtensive deformation, melt flow is expected to be either subhorizontal along the lineation in the plane of the foliation (Brown & Solar 1998a) or subvertical in discordant structures (cf. Sibson 1996; Sibson & Scott 1998). Marchildon & Brown (2003) have argued that melt flow in the DSA migmatites was along the layering parallel to the quartz-feldspar mineral aggregate lineation to interboudin partitions, shear surfaces and dykes. Upward transport of melt is inferred to have been principally through dykes or other backbone structures such as the 'pipe-like' structure formed by intersection of coeval but small dykes (Fig. 4c).
Discussion M i g m a t i t e - g r a n i t e relations
Granites emplaced in the upper crust are a necessary complement to melt-depleted lower
Fig. 10. Geometricrelationshipbetween the strike orientation of granite dykes (lighter grey ornament), the SASZ (darker grey ornament) and the crenulation cleavage associated with S3 folds.
crust (Fyfe 1973; Brown & Solar 1999); they represent a tangible result of the dissipation of strain and thermal energy. There is a tendency to separate the processes of melt generation and segregation from ascent and emplacement. However, this separation is artificial - a simple succession of events is unlikely and feedback relations are to be expected (Brown 2001a, b). Melt extraction may be a self-organized critical phenomenon in which the critical point is the threshold for melt escape from the anatectic lower crust (e.g. Handy et al. 2001; Bons et al. 2002). In actively deforming anatectic systems, the strain field that emerged under subsolidus conditions during prograde metamorphism will control the initial pattern of melt flow (Brown & Solar 1998a; Brown et al. 1999). As melt volume increases, ongoing deformation of the heterogeneous crust leads to development of an interconnected melt flow network and melt expulsion. In the DSA, the development of centimetric peritectic cordierite in interboudin partitions implies the second episode of melting was driven by decompression. The appropriate biotite dehydration-driven melting reaction involves an increase in molar volume of about 7% in crossing from the high P to low P side (Fig. 2, A or B). The dilational strain associated with progress of this reaction and increasing volume of melt, in combination with ongoing transtensive deformation and doming is likely to have motivated the interconnection of the deformation band network (cf. Grujic & Mancktelow 1998; Mandal et al. 2004). However, some parts of the remnant melt flow networks appear to exhibit scaledependent behaviour (e.g. Marchildon & Brown 2003), which conflicts with the scale-invariant behaviour expected from self-organized critical phenomena (e.g. Bons et al. 2004). It is likely that the scale-dependent behaviour is a function of intrinsic anisotropy in the system. Whether or not the extraction of melt is a self-organized critical phenomenon, the patterning of leucosome in residual lower crust provides evidence of the physical process by which this is accomplished in nature (Solar & Brown 2001; Brown 2004). Draining of melt from deformation bands and capture of small conduits by large conduits will be favoured by the lower melt pressure in the larger conduits. However, the anisotropy of stromatic leucosomes and their small thickness (millimetric to centimetric) impose a limit on effective melt transport, and in strike-slip (transpressive and transtensive) systems in particular, stromatic leucosomes cannot have been responsible for subvertical melt transport (Rutter
SYNERGY BETWEEN MELTING AND DEFORMATION 1997). Rather, in the migmatites of the DSA, stromatic leucosomes fed steeply dipping crosscutting leucosomes and granite dykes that formed synchronously with late Carboniferous (D3) dextral displacement along the SASZ. Melt flow appears to have occurred through a deformation band network at outcrop scale and in dykes at crustal scale. Dyke growth is governed primarily by the ratio of elastic to viscous response of the host rock. As viscosity contrast is reduced with increasing T and proportion of melt, the viscous response dominates, although the resulting intrusions are dyke-like. The petrographic continuity across apparently discordant contacts between granite in dykes and the leucosomes in the stromatic migmatites indicates that stromata, interboudin partitions, shear bands and dykes all contained melt during the period of this lateCarboniferous dextral transtensive deformation. Also, the dykes demonstrate that melt extraction and ascent through the crust had occurred at and below the crustal level exposed; that is, there is a flux of melt through any particular level of exposure in the source. The cumulative frequency distribution of dyke thicknesses suggests that dyke formation may be scale-invariant, which is consistent with the likelihood that dyke spacing and dyke width are controlled by the material properties at a scale larger than the migmatite anisotropy. In S Brittany, spacing of conduits is inferred to have been controlled by the strong intrinsic permeability anisotropy in layers with well-developed stromatic layering and the weak implied extrinsic permeability anisotropy between layers, which leads to variable focusing of melt that requires a larger collection zone for each conduit than would be the case for an isotropic protolith (cf. Eichhubl & Boles 2000). However, overall the limited scale of observations of dyke widths across only two orders of magnitude leaves ambiguity in this result, and scale invariance of melt extraction and ascent remains to be demonstrated convincingly in nature. The switch from ascent to emplacement may be caused by amplification of instabilities (Brown 2001a, b), where instabilities may occur within (e.g. caused by variations in permeability or magma flow rate) or surrounding the ascent column (e.g. caused by variations in the strength or state of stress). Alternatively, ascending magma may encounter a weaker layer in the crust or intersect the brittle-ductile transition zone or some other discontinuity, any of which may enable horizontal magma emplacement and pluton inflation (Roman Berdiel et al. 1995, 1997; Brown & Solar 1998b; Handy
219
et al. 2001). In the case of the DSA, magma
ascent appears to have stalled around the brittle-ductile transition zone and to have been emplaced horizontally approximately along the interface between the infrastructure and the suprastructure. In the DSA, the lower crustal unit is narrow and parallel to the SASZ, which is a major crustal-scale tectonic discontinuity, and leucogranite melt ascent and emplacement was diachronous across the DSA. To the northwest side of the SASZ older plutons are spatially related to the SASZ, which could have been a major backbone element for early melt extraction from anatectic lower crust. During emplacement, leucogranite magma expanded into the Domaine Centre-Armoricain to form plutons (Martelet et al. 2004). Along the SASZ, granites occur as steeply oriented tabular plutons. To the southwest side of the SASZ younger granites occur as thin subhorizontal tabular plutons emplaced along horizons now reactivated as extensional detachments. The system of structures observed in lower crustal migmatites is inferred to have accommodated large-scale buoyancy-driven melt transport to feed Upper Carboniferous two-mica leucogranites. The importance of the two-mica granites in focusing deformation along regional-scale shear zones and detachments has been noted by a number of authors (e.g. Strong & Hanmer 1981; Lagarde et al. 1990; Gapais et al. 1993; Brown & Dallmeyer 1996). The temporal migration of the focus of Carboniferous leucogranite magmatism from earlier ascent and emplacement spatially related to the SASZ (Strong & Hanmer 1981) to later ascent through the migmatites and emplacement at detachments between units (Gapais et al. 1993) is inferred to reflect the change in geodynamics from sinistral transpression to dextral transtension. This migration implies that early melt flow was subhorizontal toward the SASZ. In its present configuration, the SASZ does not extend downward to the Moho, but has been displaced to the NE by thrusting (Bitri et al. 2003; Martelet et al. 2004). However, images of the upper mantle based on seismic tomography show that the SASZ may be traced through the whole lithosphere (Judenherc et al. 2003). Thus, at the time of earlier Carboniferous leucogranite magmatism, it is likely that the SASZ cut through the suprasolidus lower crust and provided a channel by which melt could escape. In contrast, the synchroneity among the second melting event recorded in the DSA lower crustal migmatites, represented by the cordierite-bearing granite-leucosome network,
220
M. BROWN
and the two-mica granites, such as the Ploemeur, Quiberon, Sarzeau and Gu6rande plutons emplaced along extensional detachments at higher crustal levels (Brown & Dallmeyer 1996), suggests that there was melt transfer directly from the anatectic zone, represented by the lower crustal migmatites, via dykes with rare 'mega-dykes' representing the roots or feeders to the sites of emplacement of former plutons now lost to erosion. The inference is that Upper Carboniferous exhumation of the DSA was facilitated by melt trapped at horizons that became reactivated as extensional detachments (Gapais et al. 1993), and segregation, ascent and emplacement of crustally-derived leucogranite magma was intimately linked to regional-scale late orogenic deformation and core complex formation.
Core complex formation
The distribution of melt at depth in an orogen will influence the strain distribution in the overlying subsolidus crust (e.g. Brunet al. 1994). In the DSA, the lower crustal units are exposed in structural culminations, interpreted to be the result of transtension that allowed the middle and lower crustal units to rise up, initiating several dome-like structures (Audren 1987; Tirel et al. 2003) and imposing boudin-like structures at the crustal scale (Gapais et al. 1995). As the weak melt-beating crust of the DSA responded to the regional transtension by rising, so the migmatitic layering and fabric were rotated from subhorizontal in the lower crust to the present subvertical orientation. Brown & Dallmeyer (1996) suggested that the Carboniferous leucogranites and the formation of a core complex in the DSA are the crustal record of slab detachment, possibly triggered by the change from close-to-orthogonal sinistral transpressive to highly oblique dextral transtensive deformation. Although slab detachment may not have occurred (see discussion above), the second melting event that led to melt loss from lower crust and core complex formation may still reflect a change in the deformation regime. The trigger for the second melting episode is inferred to be the change from sinistral transpression to dextral transtension. The reasoning behind this inference is that a change from transpression to transtension that involves reversal of the sense of shear requires a switch among the principal stresses. Such a switch is inferred to induce incipient sites of dilation and boudinage, and in appropriate circumstances the small
decrease in P at spaced intervals along competent layers in extension may initiate a metamorphic reaction with positive d P / d T slope (in this case the biotite dehydration induced melting reaction involving cordierite as a peritectic product; see Fig. 2). This certainly appears to have been what has happened in the DSA migmatites, since the peritectic cordierite is located in interboudin partitions that clearly must have been sites of melting to generate the cordierite, and also sites of melt loss to preserve the cordierite. The production of small amounts of melt occurs over a very small drop in P for this particular reaction, but the effective reduction in density and rheology throughout the volume of suprasolidus orogenic crust implied by the exposed lower crustal migmatite unit in the DSA likely was sufficient to facilitate the initial rise of this unit during the early stage of transtensive deformation. This initiated a feedback relation between lower crustal doming and decompression melting, and between dome amplification and melt extraction and emplacement in developing extensional detachments. The increase in volume associated with the melting reaction in combination with doming in response to the dextral transtensive deformation drives interconnection of the deformation band network, parts of which likely are reactivated structures from the earlier melting event. The dextral shear component of the deformation may have helped to facilitate extraction of the melt generated during decompression. This process likely is terminated once sufficient melt has accumulated at the base of the Belle Ile Group to allow detachment of upper crustal units from the lower crustal unit, which enabled rapid tectonic exhumation and fast cooling of the migmatite source, and terminated reaction and froze any residual melt. Marchildon and Brown (2003) suggested that melt loss at this stage accumulated at shallower crustal levels and was emplaced along the lower contact of the Belle-Ile Group to facilitate extensional detachment of the upper crustal units. This level probably was close to the Upper Carboniferous brittle-ductile transition zone (e.g. Cagnard et al. 2004); the strength peak in the crust at this level probably represents a natural magma trap, particularly when the effects of magma pressure in relation to the regional stress field are considered (Brisbin 1986; Vigneresse et al. 1999), unless the melt begins ascent with a magmastatic head sufficient to break through this barrier. Displacement related to the detachment is recorded by the late-magmatic-early-subsolidus S-C fabrics
SYNERGY BETWEEN MELTING AND DEFORMATION present in the Quiberon, Sarzeau and Gu&ande plutons (Bouchez et al. 1981; Audren 1987; Gapais et al. 1993; Brown & Dallmeyer 1996). In contrast, at the southeast extremity of the DSA around Les Sables d'Olonne, the Upper Carboniferous deformation was accommodated by pervasive extension beneath the brittleductile transition zone and localization and core complex formation did not occur (Cagnard et al. 2004). This difference may be related to degree of melting in the lower crust, since core complex formation appears to require local weak heterogeneities in the lower crust and in the absence of such weak heterogeneities distributed thinning occurs (Brunet al. 1994). Since the pressure of metamorphism in the core complexes around Vannes and St. Nazaire is not significantly different than that at the structural base of the section at Les Sables d'Olonne, this cannot relate to differential thickening during the contractional (transpressive) phase of orogenesis but may instead relate to the bulk composition and/or fertility of the supracrustal lithologies in the lower crust here. Thus, at Les Sables d'Olonne the higher proportion of quartz-rich units in the supracrustal sequence may have limited the degree of melting, which, in turn, means that the lower crust did not provide sufficient fugitive melt to facilitate localization at the brittle-ductile transition zone and core complex formation, or the bulk composition may have been inappropriate for the cordieriteproducing biotite dehydration-driven melting reaction to occur. In summary, the following three features of the late Carboniferous evolution of the DSA support a feedback relation between transtension, a second melting event in the lower crustal unit, leucogranite ascent and emplacement, exhumation and core complex formation: (1) the decompression step in the retrograde P-T path followed by the DSA migmatites (Brown & Dallmeyer 1996; Johnson & Brown 2004); (2) the occurrence of subhorizontal tabular synkinematic plutons of leucogranite marking extensional detachments that separate high-T migmatites of the lower crustal unit from lower-T rocks of the upper crustal units (Gapais et al. 1993); and, (3) concordance of cooling ages among successively lower-T thermochronometers that indicates rapid cooling of lower crustal units and leucogranites during the late Carboniferous (Brown & Dallmeyer 1996). I thank D. Gapais and K. McCaffrey for reviews of an earlier version of this paper, T. Johnson and R. Weinberg for informal reviews, and anonymous and C. Teyssier for fornlal reviews; review comments
221
stimulated significant improvement of the paper. This work has been supported by the NSF (EAR-0003531) and the University of Maryland; Nathalie Marchildon and Barry Reno are thanked for help with preparation of the figures.
References ANTONELLINI, M. A., AYD1N, A. & POLLARD, D. D. 1994. Microstructure of deformation bands in porous sandstones at Arches National Park, Utah. Journal of Structural Geology, 16, 941-959. Au DREN, C. 1987. Evolution structurale de la Bretagne m6ridionale au Pal4ozoique. Socidtd Gdologique et Miniralogique de Bretagne Mdmoire, 31. AUDREN, C. 1990. Evolution tectonique et m4tamorphique de la chaTne varisque en Bretagne m6ridionale. Schweizerische Mineralogische und Petrographische Mittelunge, 70, 17-34. AUDREN, C. & PLAINE, J. 1986. Notice explicative de la feuille Belle-Ile-en-Mer et Iles Houat et Hoedic (Carte G6ologique de la France 1:50,000). Orldans: Editions du B.R.G.M. AtJDREN, C. & TRmOULET, C. 1993. P-T-t-deformation paths recorded by Kinzigites during diapirism in the western Variscan Belt (Golfe du Morbihan, southern Brittany, France). Journal of Metamorphic Geology, 11, 337-357. AUTRAN, A. & COGNt~, J. 1980. La zone interne de l'orog~ne varisque dans l'Ouest de la France et sa place dans le d~veloppement de la chaine hercynienne. Mdmoire BRGM, 108, 90-111. BALLI~VRE, M., PITRA, P. 8`: BOHN, M. 2003. Lawsonite growth in the epidote blueschist from the Ile de Groix (Armorican Massif, France): A potential geobarometer. Journal of Metamorphic Geology, 21, 723-735. BARRAUD, J., GARDIEN, V., ALLEMAND, P. & GRANDJEAN, P. 2004. Analogue models of meltflow networks in folding migmatites. Journal of Structural Geology, 26, 307-324. BERNARD-GRIFFITHS,J., PEUCAT, J.-J., SHEPPARD, S. & VIDAL, P. 1985. Petrogenesis of Hercynian leucogranites from the southern Armorican Massif; contribution of REE and isotopic (St, Nd, Pb and O) geochemical data to the study of source rock characteristics and ages. Earth and Planetary Science Letters, 74, 235-250. BERNARD-GRIFFITHS, J., CARPENTER, M. S. N., PEUCAT, J.-J. 8,: JAHN, B. M. 1986. Geochemical and isotopic characteristics of blueschist facies rocks from the Ile de Groix, Armorican Massif (Northwest France). Lithos, 19, 235-253. BERTHE, D., CHOUKROUNE,P. & JI~GOUZO, P. 1979. Orthogneiss, mylonite and non coaxial deformation of granites: the example of the South Armorican Shear Zone. Journal of Structural Geology, 1, 31-42. BITRI, A., BALLI~VRE, M. et al. 2003. Imagerie sismique de la zone de collision hercynienne dans le sud-est du massif Armorican (Projet Armor 2/ G6ofrance 3D). Comptes Rendus Geosciences, 335, 969-979.
222
M. BROWN
BONS, P. D. 1999. Apparent extensional structures due to volume loss. Proceedings of the Estonian Academy of Sciences, Geology, 48, 3-14. BONS, P. D., DOUGHERTY-PAGE,J. & ELBURG, M. A. 2001. Stepwise accumulation and ascent of magmas. Journal of Metamorphic Geology, 19, 627-633. BONS, P. D., ELBURG, M. A., VAN MILLIGEN, B. P., SOESOO, A. & PASSCHIER, C. W. 2002. A numerical model for self-organized criticality in melt accumulation and ascent in partially molten rocks. In: PREISS, V. P. (ed) Geoscience 2002: Expanding Horizons. Abstracts of the 16th Australian Geological Convention, Adelaide Convention Center, Adelaide, SA, Australia. July 1-5, 2002, No. 67, 206. BONS, P. D., ARNOLD, J., ELBURG, M. A., CALDER, J. & SOESOO, A. 2004. Melt accumulation from partially molten rocks. Lithos, 78, 25-42. BOSSE, V., FI~RAUD, G., BAELEVRE, M. & PEUCaT, J.-J. 2000. Petrological and geochronological constraints in the Ile de Groix blueschist (Armorican Massif, France). Basement Tectonics, 15, 63-66. BOSSE, V., BALLI~VRE,M. & VIDAL, O. 2002. Ductile thrusting recorded by the garnet isograd from the blueschist-facies metapelites of the Ile de Groix, Armorican Massif, France. Journal of Petrology, 43, 485-510. BOSSlERE, G. 1988. Evolutions chimico-min~ralogiques du grenat et de la muscovite au voisinage de l'isograde biotite-staurotide dans un mGtamorphisme prograde de type barrovien: un exemple en Vendee littorale (Massif Armoricain). Comptes Rendus de l'Acaddmie des Sciences, Paris, Sdrie //, 306, 135-140. BOUCHEZ, J.-L., GU1LLET, P. & CHEVALIER, F. 1981. Structures d'Gcoulement li~es ~ la mise en place du granite de GuGrande (Loire-Atlantique, France). Bulletin de Socidtd Gdologique de France, 7, 387-399. BmSB1N, W. C. 1986. Mechanics of pegmatite intrusion. American Mineralogist, 71, 644-651. BROWN, M. 1983. The petrogenesis of some migmatites from the Presqu'i'le de Rhuys, southern Brittany, France. In: ATHERTON, M. P. & GRIBBLE, C. D. (eds) Migmatites, Melting and Metamorphism. Shiva Publishing, Nantwich, 174- 2OO. BROWN, M. 1994. The generation, segregation, ascent and emplacement of granite magma: the migmatiteto-crustally-derived granite connection in thickened orogens. Earth Science Reviews, 36, 83-130. BROWN, M. 2001a. Crustal melting and granite magmatism: key issues. Physics and Chemistry of the Earth (A), 26, 201-212. BROWN, M. 200lb. Orogeny, migmatites and leucogranites: a review. Proceedings of the Indian Academy of Sciences (Earth and Planetar)' Science), 110, 313-336. BROWN, M. 2004. The mechanism of melt extraction from lower continental crust of orogens. Transactions of the Royal SocieO' of Edinburgh, Earth Sciences, 95, 35-48.
BROWN, M. & DALLMEYER,R. D. 1996. Rapid Variscan exhumation and the role of magma in core complex formation: southern Brittany metamorphic belt, France. Journal of Metamorphic Geology, 14, 361-379. BROWN, M. & O'BRIEN, P. J. 1997. Evolution of metamorphic belts: a changing view. In: XIANGHN, Q., ZHENDONG, Y. & HALL, H. C. (eds) Precambrian Geology and Metamorphic Petrology. Proceedings of the 30th International Geological Congress, VSP, Zeist, The Netherlands, 17, 217-231. BROWN, M. & RUSHMER, T. 1997. The role of deformation in the movement of granite melt: views from the laboratory and the field. In: HOLNESS, M. B. (ed) Deformation-Enhanced Fluid Transport in the Earth's Crust and Mantle. The Mineralogical Society Series: 8. Chapman and Hall, London, 111-144, BROWN, M. & SOLAR, G. S. 1998a. Shear zone systems and melts: Feedback relations and selforganization in orogenic belts. Journal of Structural Geology, 20, 211-227. BROWN, M. & SOLAR, G. S. 1998b. Granite ascent and emplacement during contractional deformation in convergent orogens. Journal of Structural Geology, 20, 1365-1393. BROWN, M. & SOLAR, G. S. 1999. The mechanism of ascent and emplacement of granite magma during transpression: a syntectonic granite paradigm. Tectonophysics, 312, 1-33. BROWN, M., AVERKIN, Y. A., MCLELLAN, E. L. & SAWYER, E. W. 1995. Melt segregation in migmatites. Journal of Geophysical Research - Solid Earth, 100, 15 655-15 679. BROWN, M. A., BROWN, M., CARLSON, W. D. & DENISON, C. 1999. Topology of syntectonic melt flow networks in the deep crust: inferences from three-dimensional images of leucosome geometry in migmatites. American Mineralogist, 84, 1793-1818. BRUN, J.-P. & BURG, J.-P. 1982. Combined thrusting and wrenching in the Ibero-Armorican arc: a corner effect during continental collision. Earth and Planetar)' Science Letters, 61, 319-332. BRUN, J.-I., SOKOUTIS, D. & VANDENDRIESSCHE,J. 1994. Analog modeling of detachment fault systems and core complexes. Geology, 22, 319-322. BURG, J.-P., BALI~, P., BRUN, J.-P. & GIRARDEAU, J. 1987. Stretching lineation and transport direction in the Ibero-Armorican arc during the SiluroDevonian collision. Geodynamica Acta, 1, 71-87. BURG, J.-P., VAN DEN DRIESSCHE, J. & BRUN, J.-P. 1994. Syn- to post-thickening extension in the Variscan Belt of western Europe: modes and structural consequences. Gdologie de la France, 1994, 33-51. CAGNARD, F., GAPAIS, D., BRUN, J.-P., GUMIAUX, C. & VAN DEN DRIESSCHE, J. 2004. Late pervasive crustal-scale extension in the south Armorica Hercynian belt (Vendee, France). Journal of Structural Geology, 26, 435-440. CARPENA, J., CHAILLOU, D., CHAMBAUDET, A. & POUPEAU, G. 1979. Fission track geochronology
SYNERGY BETWEEN MELTING AND DEFORMATION of the Hercynian Platform in France. In: Report of the lOth Annual Conference on Solid State Nuclear Track Detectors. Space Biophysics of the Parliament of the Council of Europe, Lyon. CARTIER, C., FAURE, M. & LARDEUX, H. 2002. Discussion on 'a new model for the Hercynian orogen of Gondwana France and Iberia' by D. Shelley & G. Bossi~re, Journal of Structural Geology 2000, 22, 757-776. Journal of Structural Geology, 24, 1659-1663. CLEMENS, J. D. 1998. Observations on the origins and ascent mechanisms of granitic magmas. Journal of the Geological Society, 155, 843-851. COGNE, J. 1960. Schistes cristallins et granites en Bretagne mGridionale: le domaine de l'Anticlinal de Cornouaille. Mimoire du Service de la Carte Giologique de France, 1, 382. DEWEY, J. & BURKE, K. C. 1973. Tibetan, Variscan and Precamhrian basement reactivation: products of continental collision. Journal of Geology, 81, 683 -692. DIAS, R. & RIBEIRO, A. 1995. The Ibero-Armorican arc: A collision effect against an irregular continent? Tectonophysics, 246, 113-128. DJRO, S. C., TR1BOULET, C. & AUDREN, C. 1989. Les chemins pression- temperature- temps-dGformation-espace ( P - T - t - d - e ) dans les micaschistes associGs aux schistes bleus de l'~le de Groix, Bretagne M~ridionale, France. Schweizerische Mineralogische und Petrographische Mitteilungen, 69, 73-90. Du BERNARD, X., E1CHHUBL, P. & AYDIN, A. 2002. Dilation bands: a new form of localized failure in granular media. Geophysical Research Letters, 29, DOI: 10.1029/2002GL015966. EICHHUBL, P. & AYDIN, A. 2003. Ductile openingmode fracture by pore growth and coalescence during combustion alteration of siliceous mudstone. Journal of Structural Geology, 25, 121 - 134. EICHHUBL, P. & BOLES, J. R. 2000. Focused fluid flow along faults in the Monterey Formation, Coastal California. Geological Society of American Bulletin, 112, 1667-1679. EICHHUBL, P., AYDIN, A. & LORE, J. 2001. Openingmode fracture in siliceous mudstone at high homologous temperature - effective of surface forces. Geophysical Research Letters, 28, 1299-1302. FAURE, M. & PONS, J. 1991. Crustal thinning recorded by the shape of the Namurian-Westphalian leucogranite in the Variscan Belt of the Northwest Massif Central, France. Geology, 7, 730-733. FAURE, M., LELOIX, C. & ROIG, J.-Y. 1997. L'E~volution polycyclique de la chaine hercynienne. Bulletin de la Sociiti Gdologique de France, Huitikme Sirie, 168, 695-705. FYFE, W. S. 1973. The granulite facies, partial melting and the Archaean crust. Philosophical Transactions of the Royal Society of London, 273A, 457-461. GAPA1S, D. 1989. Shear structures within deformed granites - mechanical and thermal indicators. Geology, 17 (12), 1144-1147. GAPAIS, D., LAGARDE, J.-L., LE CORRE, C., AUDREN, C., JIEGOUZO, P., CASAS SAINZ, A. &
223
VAN DEN DRIESSCHE, J. 1993. La zone de cisaillement de Quiberon: tGmoin d'extension de la cha~ne varisque en Bretagne mGridionale au CarbonifGre. Comptes Rendus de l'Academie Sciences Paris, Serie 2, 316, 1123-1129. GAPAIS, D., VAN DEN DRIESSCHE, J., BRUN, J.-P. & RICHER, C. 1995. Pervasive ductile spreading and boudinage in extending orogens; evidence from the French Variscan Belt. TERRA Abstracts, 7, 121. GEOFFROY, L. 1993. Tectonique tardi-varisque en failles normales ductiles en VendGe Littorale, Massif Armoricain. Cornptes Rendus des S~ances de l'Acaddmie des Sciences, 317, 1237-1243. GERDES, A., WORNER, G. & HENK, A. 2000. Postcollisional granite generation and H T - L P metamorphism by radiogenic heating: the Variscan South Bohemian Batholith. Journal of the Geological Society, London, 157, 577-587. GEEASON, G. C., BRUCE, V. & GREEN, H. W. 1999. Experimental investigation of melt topology in partially melted quartzo-feldspathic aggregates under hydrostatic and non-hydrostatic stress. Journal of Metamorphic Geology, 17, 705-722. GODARD, G. 1988. Petrology of some eclogites in the Hercynides: The eclogites from the southern Armorican Massif, France. In: SMITH, D. C. (ed) Eclogites and Eclogite-Facies Rocks. Elsevier, Amsterdam, 451-519. GODARD, G. 2001. Les Essarts eclogite-bearing metamorphic complex (VendGe, Southern Armorican Massif, France). Gdologie de la France, 2001 (1-2), 19-51 & 2 maps. GouJou, J. C. 1992. Analyse pGtro-structurale dans un avant-pays m~tamorphique: influence du plutonisme tardi-orogGnique varisque sur l'encaissant 6pi fi mGsozonal de VendGe. Documents du BRGM, 216. GRUJIC, D. & MANCKTELOW,N. S. 1998. Melt-bearing shear zones: analogue experiments and comparison with examples from southern Madagascar. Journal of Structural Geology, 20, 673-680. GUERNINA, S. & SAWYER, E. W. 2003. Large-scale melt-depletion in granulite terranes: An example from the Archean Ashuanipi Subprovince of Quebec. Journal of Metamorphic Geology, 21, 181-201. GUIRAUD, M., BURG, J.-P. & POWELL, R. 1987. Evidence for a Variscan suture zone in the VendGe, France: a petrological study of blueschist facies rocks from Bois de Cen& Journal of Metamorphic Geology, 5, 225-237. GUMIAUX, G., JUDENHERC, S., BRUN, J.-P., GAPAIS, D., GRANET, M. & POUPINET, G. 2004. Restoration of lithosphere-scale wrenching from integrated structural and tomographic data (Hercynian belt of western France). Geology, 32, 333 -336. HAND, M. & DIRKS, P. H. G. M. 1992. The influence of deformation on the formation of axial-planar leucosomes and the segregation of small melt bodies within the migmatitic Napperby Gneiss, central Australia. Journal of Structural Geology, 14, 591-604.
224
M. BROWN
HANDY, M. R., MULCH, A., ROSENAU, N. & ROSENBERG, C. L. 2001. The role of fault zones and melts as agents of weakening, hardening and differentiation of the continental crust: a synthesis. In: HOLDSWORTH, R. E., MAGLOUGHLIN, J., KNIPE, R. J., STRACHAN, R. A. & SEARLE, R. C. (eds). The Nature and Tectonic Significance of Fault Zone Weakening. Geological Society Special Publications, 186, 305-332. HANMER, S. 1997. Shear zone reactivation at granulite facies: The importance of plutons in the localization of viscous flow. Journal of the Geological Society, 154, 111 - 116. HILL, J., BALDWIN, S. L. &LlSTER, G. S. 1995. Magmatism as an essential driving force for formation of active metamorphic core complexes in eastern Papua New Guinea. Journal of Geophysical Research, 100, 10 441-10 451. HODC,ES, K. V. 2000. Tectonics of the Himalayas and southern Tibet from two prospectives. Geological Society of America Bulletin, 112, 324-350. HOLmSTER, L. S. 1993. The role of melt in the uplift and exhumation of orogenic belts. Chemical Geology, 108, 31-48. HOLYOKE, III, C. W. & RUSHMER, T. 2002. An experimental study of grain scale melt segregation mechanisms in two common crustal rock types. Journal of Metamorphic Geology, 20, 493-512. Jr~GOUZO, P. 1980. The South Armorican shear zone. Journal o[ Structural Geology, 2, 39-47. JEGOUZO, P. & ROSSELLO, E. A. 1988. La branch6 nord du cisaillement Sud-Armoricain (France): un essai d'6valuation du d6placement par l'analyse des mylonites. Comptes Rendus de l'Academie Sciences Paris, Serie 2,307, 1825-1831. JOHNSON, T. & BROWN, M. 2004. Quantitative constraints on metamorphism in the Variscides of southern Brittany - a complementary pseudosection approach. Journal of Petrology, 45, 1237-1259. JONES, K. A. 1988. The metamorphic petrology of the southern Brittany Migmatite Belt, France. PhD dissertation, Kingston Polytechnic University. JONES, K. A. 1991. Paleozoic continental margin tectonics of southern Armorica. Journal of the Geological Society, London, 148, 55-64. JONES, K. A. & BROWN, M. 1989. The metamorphic evolution of the southern Brittany Migmatite Belt. In: DALY, J. S., CLIFF, R. A. & YARDLEY, B. W. D. (eds) Evolution of Metamorphic Belts. Geological Society Special Publications, 43, 501-505. JONES, K. A. & BROWN, M. 1990. High-temperature 'clockwise' P-T paths and melting in the development of regional migmatites: an example from southern Brittany, France. Journal of Metamorphic Geology, 8, 551-578. JUDENHERC, S., GRANET, M., BRUN, J.-P., POUP1NET, G., PLOMEROVA, J., MOCQUET, A. & ACHAUER, U. 2002. Images of lithospheric heterogeneities in the Armorican segment of the Hercynian Range in France. Tectonophysics, 358, 121-143.
JUDENHERC, S., GRANET, M., BRUN, J.-P. & POUPINET, G. 2003. The Hercynian collision in the Armorican Massif; evidence of different lithospheric domains inferred from seismic tomography and anisotropy. Bulletin de la Socidtd Gdologique de France, 174, 45-57. LAGARDE, J.-L., BRUN, J.-P. & GAPAIS, D. 1990. Formation des plutons granitiques par injections et expansion lat&ale dans leur site de raise en place: une alternative au diapirisme en domaine ~pizonal. Comptes Rendus des Sdances de l'Acadimie des Sciences, 310, 1109-1114. LE HEI3EL, F. 2002, Ddformation continentale et histoire des fluides au cours d'un cycle subduction, exhumation, extension. Exemple des porphyroi'des sud-armoricains. Unpublished PhD thesis, University of Rennes I. LE HI2BEL, F., GAPAIS, D., FOURCADE, S. &; CAPDEVILA, R. 2002a. Fluid-assisted large strains in a crustal-scale ddcollement (Hercynian belt of south Brittany, France). Geological Society Special Publications, 200, 85-101. LE HI~BEL,F., VIDAL, 0., KIENAST,J. R. ~; GAPAIS, D. 2002b. Les 'Porphyroides' de Bretagne m6ridionale: une unit6 de HP-BT dans la chaine hercynienne. Comptes Rendus G&)science, 334, 205-211. MANDAL, N., MISRA, S. & SAMANTA, S. K. 2004. Role of weak flaws in nucleation of shear zones: an experimental and theoretical study. Journal of Structural Geology, 26, 1391 - 1400. MARCHILDON, N. & BROWN, M. 2001. Melt segregation in late-tectonic anatectic migmatites: an example from the Onawa contact aureole, Maine, U.S.A. Physics and Chemistry of the Earth, 26, 225 -229. MARCHILDON, N. &; BROWN, M. 2002. Grain-scale melt distribution in two contact aureole rocks: implications for controls on melt localization and deformation. Journal of Metamorphic Geology, 20, 381-396. MARCHILDON, N. & BROWN, M. 2003. Spatial distribution of melt-bearing structures in anatectic rocks from southern Brittany, France: implications for melt-transfer at grain-scale to orogen-scale. Tectonophysics, 364, 215-235. MARTELET, G., CALCAGNO, P., GUMIAUX, C., TRUFFERT, C., BITRI, A., GAPAIS, D. & BRUN, J.-P. 2004. Integrated 3D geophysical and geological modelling of the Hercynian Suture Zone in the Champtoceaux area (south Brittany, France). Tectonophysics, 382, 117-128. MATTE, P. 1986. Tectonics and plate tectonic model for the Variscan belt of Europe. Tectonophysics, 126, 329-374. MATTE, P. 1991. Accretionary history and crustal evolution of the Variscan belt in Western Europe. Tectonophysics, 196, 309-337. MATTE, P. 2001. The Variscan collage and orogeny (480-290 Ma) and the definition of the Armorica microplate: tectonic approach. Terra Nova, 13, 122-128. MCKENZIE, D. P & JACKSON, J. 2002. Conditions for flow in the continental crust. Tectonics, 21, 1055, doi: 10.1029/2002TC001394.
SYNERGY BETWEEN MELTING AND DEFORMATION MOLLEMA, P. N. & ANTONELLINI,M. A. 1996. Compaction bands; a structural analog for anti-mode I cracks in Aeolian sandstone. Tectonophysics, 267, 209-228. OLIVER, N. H. S. & BARR, T. D. 1997. The geometry and evolution of magma pathways through migmatites of the Halls Creek Orogen, Western Australia. Mineralogical Magazine, 61, 3-14. PARSONS, T. & THOMPSON, G. A. 1993. Does magmatism influence low-angle normal faulting? Geology, 21, 247-250. PETFORD, N. & KOENDERS,M. A. 1998. Self-organization and fracture connectivity in rapidly heated continental crust. Journal of Structural Geology, 20, 1425-1434. PEUCAT, J.-J. 1983. Grochronologie des roches mrtamorphiques (Rb-Sr et U-Pb). Mdmoire de la Socidtd Gdologique et Mindralogique de Bretagne, 28, 1-158. PIN, C. & PEUCAT, J.-J. 1986. Ages des ~pisodes de m6tamorphisme pa16ozoi'que dans le Massif central et le Massif Armoficain. Bulletin de Socidtd Gdologique de France, 8, 461-469. QUINQU1S, H. & CHOUKROUNE, P. 1981. Les schistes bleus de File de Groix dans la cha~ne hercynienne: implications cinrmatiques. Bulletin de la Socidtd Gdologique de France, 23, 409-418. REGENAUER-LIEB, K. 1999. Dilatant plasticity applied to Alpine collision: Ductile void growth in the intraplate area beneath the Eifel volcanic field. Journal of Geodynamics, 27, 1-21. REY, P., BURG, J. P. & CASEY, M. 1997. The Scandinavian Caledonides and their relationship to the Variscan belt. In: BURG, J. P. & FORD, M. (eds) Orogeny Through Time. Special Publications Geological Society, 121, 179-200. ROMAN BERDIEL, T., GAPAIS, D. & BRUN, J.-P. 1995. Analog models of laccolith formation. Journal of Structural Geology, 17, 1337-1346. ROMAN BERDIEL, T., GAPAIS, D. & BRUN, J.-P. 1997. Granite intrusion along strike-slip zones in experiment and nature. American Journal of Science, 297, 651-678. RUBATTO, D., WILLIAMS, I. S. & BUICK, I. S. 2001. Zircon and monazite response to prograde metamorphism in the Reynolds Range, central Australia. Contributions to Mineralogy and Petrology, 140, 458-468. RUTTER, E. H. 1997. The influence of deformation on the extraction of crustal melts: a consideration of the role of melt-assisted granular flow. In: HOLNESS, M. B. (ed) Deformation-Enhanced Fluid Transport in the Earth's Crust and Mantle. Mineralogical Society Series, 8, 82-110. SAWYER, E. W. 2001. Melt segregation in the continental crust: distribution and movement of melt in anatectic rocks. Journal of Metamorphic Geology, 19, 291-309. SCHULTZ, B., TRIBOULET, C., AUDREN, C., PFEIFER, H.-R. & GILG, A. 2001. Two-stage prograde and retrograde Variscan metamorphism of glaucophane-eclogites, blueschists and greenschists from the Ile de Groix (Brittany, France). International Journal of Earth Sciences, 90, 871-889.
225
SCHULTZ, B., AUDREN, C. & TRIBOULET, C. 2002. Oxygen isotope record of fluid-rock-SiO2 interaction during Variscan progressive deformation and quartz veining in the meta-volcano sediments of Belle-Ile (Southern Brittany). Journal of Structural Geology, 24, 1281-1297. SHELLEY, O. & BOSSII~RE,G. 1999. Ile de Groix: retrogression and structural developments in an extensional rrgime. Journal of Structural Geology, 21, 1441-1455. SHELLEY, D. & BOSSII~RE, G. 2000. A new model for the Hercynian Orogen of Gondwanan France and Iberia, Journal of Structural Geology, 22, 757 -776. SmSON, R. H. 1996. Structural permeability of fluiddriven fault - fracture meshes. Journal of Structural Geology, 18, 1031 - 1042. SIBSON, R. H. & SCOTT, J. 1998. Stress/fault controls on the containment and release of over pressured fluids: examples from gold-quartz vein systems in Juneau, Alaska; Victoria, Australia and Otago, New Zealand. Ore Geology Reviews, 13, 293-306. SOLAR, G. S. & BROWN, M. 2001. Petrogenesis of migmatites in Maine, USA: possible source of peraluminous leucogranite in plutons? Journal of Petrology, 42, 789-823. STRONG, D. F. & HANMER, S. K. 1981. The leucogranites of southern Brittany; origin by faulting, frictional heating, fluid flux and fractional melting. Canadian Mineralogist, 19, 163-176. TEYSSIER, C. & WHITNEY, D. 2002. Gneiss domes and orogeny. Geology, 30, 1139-1142. TIREL, C., BRUN, J.-P. & BUROV, E. 2003. Thermomechanical modelling of extensional gneiss domes. Deformation Mechanisms, Rheology and Tectonics. G~osciences Rennes, St. Malo, France, April 14-16, 2003, Abstract Volume, 160. TRmOULET, C. 1974. Les glaucophanites et roches associres de File de Groix (Morbihan, France): 6tude min~ralogique et prtrogrn6tique. Contributions to Mineralogy and Petrology, 45, 65-90. TRIBOULET, C. 1991. Etude gdothermo-baromrtrique compar6 des schistes bleus palrozo'iques de l'Ouest de la France (Ile de Groix, Bretagne Mrridionale et Bois de Cent, Vendre). Comptes Rendus des Sdances de l'Acaddmie des Sciences, 312, 1163-1168. TRIBOULET, C. & AUDREN, C. 1985. Continuous reactions between biotite, garnet, staurolite, kyanitesillimanite-andalusite and P-T-time-deformation path in micaschists from the estuary of the river Vilaine, South Brittany, France. Journal of Metamorphic Geology, 3, 91-105. TRIBOULET, C. 8,: AUDREN, C. 1988. Controls on P-T-t deformation path from amphiboles during progressive metamorphism with basic rocks (estuary of the river Vilaine, South Brittany, France). Journal of Metamorphic Geology, 6, 117-133. TRUFFERT, C., GUMIAUX, C. ET AL. 2001. Lev~ g~ophysique arroport6 dans le Sud-Est du Massif armoricain (programme GroFrance3D Armor2). Magn~tisme et Radiom~trie spectrale. Comptes
226
M. BROWN
Rendus des Sdances de l'Acaddmie des Sciences, 333, 263-270. VAN DE ZEDDE, D. M. A. & WORTEL, M. J. R. 2001. Shallow slab detachment as a transient source of heat at midlithospheric depths. Tectonics, 20, 86-882. VAN DER MOLEN, I. & PATERSON, M. S. 1979. Experimental deformation of partially-melted granite. Contributions to Mineralogy and Petrology, 70, 299-318. VANDERHAEGHE, 0. & TEYSSIER, C. 2001. Partial melting and flow of orogens. Tectonophysics, 342, 451-472.
VAtJCHEZ, A., MAILLET, D. & SOUGY, J. 1987. Strain and deformation mechanisms in the Variscan nappes of Vend6e, South Brittany, France. Journal of Structural Geology, 9, 31-40. VIDAL, P. 1973. Premieres donn~es g6ochronologiques sur les granites hercyniens du Sud du Massif Armoricain. Bulletin de Soci~td Giologique de France, 7, 239-245. VIGNERESSE, J.-L., TIKOFF, B. & AMI~GLIO, L. 1999. Modification of the regional stress field by magma intrusion and formation of tabular granitic plutons. Tectonophysics, 302, 203-224.
A critical assessment of the tectono-thermal memory of rocks and definition of tectono-metamorphic units: evidence from fabric and degree of metamorphic transformations M. IOLE SPALLA 1'2, MICHELE ZUCALI 1, SILVIA DI PAOLA 1 & GUIDO GOSSO 1'2
1Dipartimento di Scienze della Terra 'A. Desio', Universitdt di Milano, Via Mangiagalli 34, 20133 Milano, Italy (e-mails:
[email protected],
[email protected],
[email protected],
[email protected]) 2C.N.R. - IDPA, Via Mangiagalli 34, 1-20133 Milano, Italy Abstract: A correlation procedure of scattered tectonic and metamorphic imprints in the reactivated crust is elaborated from recent analytical work in three Alpine metamorphic complexes. It consists of: interpretation of the time-sequence of tectonic fabrics and test of their kinematic coherence; determination of paragenetic compatibility among the mineralogical support of mesoscopic fabrics; cross-validation of mineral transformation overprints; construction of P-T-d-t paths using a time-sequence of parageneses. The representation of structural and metamorphic information conveys the full tectono-metamorphic history on maps displaying combined tectonic and metamorphic effects. Shape and size definition of metamorphic units, now individuated mainly using their lithological homogeneity and dominant metamorphic imprint, is improved. The analysis of interactions between fabric and metamorphic imprint distributions, proposed in three Alpine examples, shows that the dominant metamorphic imprint does not coincide with Tmax-Prmaxof each inferred P-T-d-t loop; the dominant metamorphic imprint is that given by the mineralogical support of the most pervasive fabric. Different metamorphic imprints may dominate in adjacent areas of a single tectono-metamorphic unit (TMU), or equivalent metamorphic imprints may dominate in different TMUs. Therefore, lithostratigraphic setting and dominant metamorphic imprint are inefficient to contour TMUs in terrains with polyphase deformation and metamorphism, without considering multiscale heterogeneity of superposed synmetamorphic fabrics.
Coupling and decoupling of crustal slices, with their associated size variation, work in competition within subduction-collision zones to form the tectonic units of metamorphic belts. Since their structural and metamorphic evolutions are tracers of their transit throughout different levels of the lithosphere, the definition of their contours may be improved to individualize the volumes carrying distinct structural and metamorphic histories (tectono-metamorphic unit = TMU). Size definition of a TMU, and careful reconstruction of its tectono-thermal evolution, are critical to infer geological processes as (i) tectonic erosion or accretion at the trench margins, (ii) continental collision or deep subduction of continental crust (ablative subduction), (iii) exhumation velocity variation and its influence on rapid and effective meta-stabilization of HP- and UHP-LT assemblages (e.g. England & Thompson 1984; Lallemand 1999; Tao & O'Connel 1992; Cloos 1993; Ernst 2001).
A reliable assessment of the effect of these different processes requires a careful examination of the tectono-metamorphic memory of units traditionally distinguished in tectonically reactivated crust and mountain belts; therefore an analytical procedure is presented, aiming to individuate portions of crust that underwent an internally coherent tectonic and metamorphic evolution. At present, in spite of remarkable progress of analytical techniques in structural geology and petrology (e.g. Passchier et al. 1990; Brown 2001), the recognition of specific P-T re-equilibration stages of successive groups of structures remains a problem because the extent, degree and timing of metamorphic re-equilibrations and fabric changes are often inadequately known to delimit TMUs. Even in relatively well-explored belts, as the European Alps, metamorphic complexes have been contoured on the basis of their lithological homogeneity
From: GAPAIS,D., BRUN,J. P. & COBBOLD,P. R. (eds) 2005. Deformation Mechanisms, Rheology and Tectonics:from Minerals to the Lithosphere. Geological Society, London, Special Publications, 243, 227-247. 0305-8719/05/$15.00
9 The Geological Society of London 2005.
228
M. IOLE SPALLA ET AL.
(suggesting for example continental or oceanic affinities) or of their dominant metamorphic imprint. On the other hand, recent pressure (P)-temperature (7) paths reconstructions reveal different sequences of metamorphic overprints within a single nappe (or complex; e.g. Pognante 1991; Spalla et al. 1996). In this light, rocks recording the maximum number of re-equilibration steps (i.e. rocks with long memory) are fundamental to reconstruct exhaustively the structural and metamorphic evolution. Fortunately, deformation and metamorphic re-equilibration are heterogeneous processes and rocks may retain a valuable memory that permits the preservation of relics, even at very high T. On the other hand, these heterogeneities make spatial and chronological correlations rather difficult. Consequently, setting up a correlation method that reconstructs well-constrained successions of structural and metamorphic re-equilibration stages, hidden as scanty traces in the tectonothermal memory, is fundamental to separate crustal volumes that underwent significantly different deformation and metamorphic histories. T is currently accepted as the main driving force of reaction kinetics, as suggested by the Arrhenius exponential relation, and consequently the dominant metamorphic imprint of a metamorphic unit is admitted to have been imposed when rocks reached conditions of Tmax-PTmax. In these models (e.g. England & Richardson 1977; Spear et al. 1984) the different T,.... PTmax conditions, recorded in adjacent crustal volumes showing different dominant metamorphic imprints, are not aligned along the same geotherm, but locate on the P - T space the 'metamorphic field gradient'. Tma• culminations of each P - T loop are therefore diachronous (implying different exhumation times) and are considered useful to identify different metamorphic units and to assess their size. In this view, the catalysing effect of grainscale deformation on metamorphic reaction rates is completely disregarded, though the effect of strain on metamorphic recrystallization be well documented at several scales, in many orogenic belts and in various metamorphic environments (e.g. Lardeaux et al. 1982; Mcrk 1985; Pognante 1985; Koons 1986; Austrheim 1990; Rubie 1990; Spalla 1993). For this reason, we take into account the role of fabric evolution on reaction progress to individuate rock volumes characterized by a potentially 'long tectono-thermal memory' that may change throughout adjacent volumes as a function of deformation heterogeneity. Therefore, we will compare the pressure-temperature-relative time of deformation paths (P-T-d-t paths of Spalla
1993; Johnson & Vernon 1995; Passchier & Trouw 1996), reconstructed in three examples from the Western and Central Austroalpine and Central Southalpine domains of the Alps. P-T-d-t paths were inferred using P - T estimates obtained from nfineral assemblages marking superposed and kinematically incompatible tectonic fabrics. They were represented on P - T diagrams by chronologically ordered re-equilibration steps corresponding to D1, D2 . . . . Dn deformation stages. The three examples were chosen in domains that underwent polyphase or polyphase and polycyclic tectono-metamorphic evolution during pre-Alpine and Alpine times, resulting from different geodynamic settings (subduction and continental collision or lithospheric extension) as suggested by strongly contrasted thermal regimes, from HP-LT to HT-LP. We qualitatively evaluate, on maps of structural versus metamorphic re-equilibration, the interaction between distribution of highly evolved fabrics and pervasive growth of metamorphic minerals, to point out the dependence of the degree of metamorphic transformation on deformation gradients. Such a mapping technique improves the structural mapping by Johnson & Duncan (1992) or Connors & Lister (1995) and enables speculation on the physical significance of the metamorphic field gradient and on its use for the definition of TMU boundaries. In addition, TMU boundaries are compared with those of units drawn from lithostratigraphic criteria.
Correlation tools The identification of a TMU requires individuating rock volumes with coherent tectonometamorphic memory and this needs an efficient correlation method. Correlation of structural and metamorphic re-equilibration stages between adjacent rock volumes, in terrains that experienced complex structural and metamorphic evolutions, needs a joint use of different tools: correlation of superposed fabric elements on a foliation trajectory map; microstructural analysis to infer stable mineral assemblages marking superposed fabrics; absolute age data (Turner & Weiss 1963; Park 1969; Hobbs et al. 1976; Van Roermund et al. 1979; Williams 1985; Passchier et al. 1990; Johnson & Duncan 1992; Johnson & Vernon 1995; Spalla et al. 2000; Di Vincenzo & Palmeri 2001). Deformation heterogeneity makes structural correlations difficult, if exclusively based on classical geometric criteria, because it generates strong variations in deformation style across
T E C T O N O - T H E R M A L M E M O R Y OF ROCKS
adjacent rock volumes (e.g. from undeformed to mylonitic textures). The analytical procedure (Fig. 1), adopted to improve correlations, comprises the following (1)
construction of a map of the total deformation field, in which configuration of lithostratigraphy and all the fabric elements is reported; test of kinematic compatibilities between all structures and their superposition sequence, to separate successive deformation imprints; recognition of mineral compatibilities (assemblages) among the mineralogical support of successive planar and/or linear fabrics; cross-control of superposed mineral transformations in the superposed fabric sequences of adjacent heterogeneously deformed volumes.
(2)
(3)
(4)
In accordance, all the structural-petrographic maps presented in this contribution contain
7
. . . . . . . . . . . . . . . . . . .
.
229
grids of superposed foliation trajectories and indicate the associated metamorphic conditions and the fabric gradients. Foliation trajectory symbols represent interpolations of seams of foliation traces containing intbrmation on the relative chronology of structural imprints (Williams 1971; Myers 1978; Gosso et al. 1983; Johnson & Duncan 1992; Connors & Lister 1995). To estimate the degree of fabric evolution, the degree of grain-scale reorganization of the prevailing planar fabric may be regionally evaluated, either with reference to six stages of the decrenulation model (Fig. 2; Bell & Rubenach 1983: Passchier & Trouw 1996), or by differentiating areas with undeformed, normally foliated or mylonitic fabrics. In the three examples, both pre-Alpine and Alpine tectonic fabrics reach variable degrees of evolution, up to stage 5 - 6 (Bell & Rubenach 1983) in metapelites or from undeformed to mylonitic in metaintrusives. The area extent of a new differentiated foliation (evolutionary stages 4 to 6) was evaluated for each successive group of structures, together with the modal amount of associated metamorphic
t
. . . . . . .
ow to delimit tectonic t ~ lithostratiaraphy~ J ~uHnitswithin metamorphicF~L ~ - -
I
! ~ |
~
terrains? ~ ~
S~qtlellC~ OT ~IID'@.F'DO_q~
!
| !
| ~-[-5-~-7imprints
. . . . . . .
~ ~ oorlllflLlrlI
.|~ ~
| aeTorI~GTIOR
.
I
~metamorphic ]| I [~ ~ ~
.
.
.
.
~
" ~
l ~
'
not sufficient because
I
*
~
portions
l
of
the same may follow
lithostratigraphic unit
different lithostratigraphfc units may foflowthe same tectonic evolution; a renewed lithostratigraphy may arise
fro~stroi~ocolis~tio~(t~ansposition)
| 1) to evaulate the role af fabric gradients on reaction progress for individuating rocks with tong structural metamorphic~ I m~mo~i film whilh struclural and metamorphic evolutions may be reconstructed starting from detailed maps l | 2) to contour rock volumes recording coherent sequencesof tectono metamorphic imprints (: coherent tectano metamorphic~ g tectonometamorphic un ~t
Lm .e m~ 1
ll
1
~
~
.
.
.
.
.
.
1
. . . . .
~
....
~
~
. . . . . . . . .
Fig. 1. Methodology for tectonic analysis in the metamorphic crust. The main steps of the analytical procedure for an improved definition of tectonic units and metamorphic complexes are reported in the lower box and explained in the text.
230
M. IOLE SPALLA ET AL.
2
Fig. 2, Stages 1 to 6 of fabric evolution during the superposed foliation development, redrawn after Bell & Rubenach (1983). 1, original foliation; 2, crenulation; 3, crenulation and incipient differentiation; 4, differentiated crenulation cleavage; 5, obliteration of relict crenulation in Q-domains; 6, new continuous foliation.
assemblages, with the aim to display on a synthetic map the relationships between fabric evolution and degree of metamorphic transformation. Maps of heterogeneity of deformation history may be drawn to further constrain the tectono-metamorphic evolution. Finally, specific metamorphic conditions of successive deformation stages were inferred from the compatibility of mineral assemblages forming the mineralogical support of fabric elements coeval in different bulk compositions. In summary, this representation technique directly conveys the inferred P-T-d-t evolution(s), and visualizes rock memory variations, offering an immediate insight on TMU contours.
The example of the polycyclic basement of the Sesia-Lanzo Zone Geological setting
In the axial portion of the Western Alps, between the Penninic Front and the Insubric Lineament (PF and IL in Fig. 3), an HP-LT metamorphic imprint widely affects the pre-Alpine continental crust. Austroalpine continental fragments, sandwiched with slices of meta-ophiolites, are interpreted as exhumed remnants of the southern plate margin that was subducted during Alpine convergence. The Sesia-Lanzo Zone (SLZ; Fig. 4) is the widest portion of Austroalpine continental crust recording early Alpine eclogitefacies metamorphism. It has been subdivided into a lower (Gneiss Minuti Complex = GMC; Eclogitic Micaschists Complex = EMC) and an
upper element (II Dioritic-Kinzigitic Zone-IIDK; e.g. Dal Piaz et al. 1972; Compagnoni et al. 1977; Pognante et al. 1987). The upper unit records a high-pressure blueschist metamorphic imprint and its mylonitic contact with the lower unit developed under eclogite or blueschist facies conditions (Lardeaux et al. 1982; Pognante et al. 1987). In the lower unit, the Alpine polyphase deformation developed during a LT eclogite imprint, followed by a blueschist re-equilibration and by a low-pressure greenschist facies overprint (e.g. Lardeaux et at. 1982; Vuichard & Ball~vre 1988; Castelli 1991; Pognante 1991 and references therein). The HP-LT metamorphic imprint is early Alpine (130-70Ma; e.g. Oberhaensli et al. 1985; Stoeckhert et al. 1986; Hunziker et al. 1992; Rubatto et al. 1999). The GMC and EMC, both pervasively eclogitized, strongly differ in the volume percentage of greenschist re-equilibration. The GMC represents the continentocean tectonic boundary and is dominantly re-equilibrated under greenschist facies conditions generally associated with mylonitic textures (Sttinitz 1989; Spalla et al. 1991). The EMC constitutes the innermost part of the SLZ and its greenschist facies overprint is confined to discrete shear zones, more pervasively developed towards its inner boundary with the Southern Alps. Contrasting dominance of HP or greenschist metamorphic imprints in EMC and GMC, respectively, led Sttinitz (1989) and Spalla et al. (1991) to the still debated interpretation that the two complexes correspond to a large-scale strain partitioning effect, taking place during greenschist facies re-equilibration. The peculiar low T I P Alpine ratio, during the eclogitic peak (c. 10 ~ -]) and during the exhumation path (_